yusainy, cleoputri (2013) overcoming aggression: musing on...
TRANSCRIPT
Yusainy, Cleoputri (2013) Overcoming aggression: musing on mindfulness and self-control. PhD thesis, University of Nottingham.
Access from the University of Nottingham repository: http://eprints.nottingham.ac.uk/13467/1/Cleo_Yusainy_Overcoming_aggression.pdf
Copyright and reuse:
The Nottingham ePrints service makes this work by researchers of the University of Nottingham available open access under the following conditions.
This article is made available under the University of Nottingham End User licence and may be reused according to the conditions of the licence. For more details see: http://eprints.nottingham.ac.uk/end_user_agreement.pdf
For more information, please contact [email protected]
OVERCOMING AGGRESSION:
MUSING ON MINDFULNESS AND SELF-CONTROL
CLEOPUTRI AL YUSAINY
Thesis submitted to the University of Nottinghamfor the degree of Doctor of Philosophy
JULY 2013
ii
ABSTRACT
Overcoming aggression: Musing on mindfulness and self-control
The ability to restrain oneself from acting on aggressive impulses is arguably a crucial
aspect of human functioning and interaction. Yet growing evidence in the literature
suggests that people’s self-control resources may be limited and, at times, self-
controlled regulation could even increase the association between aggressive
triggers and aggressive behaviour. As an alternative, mindfulness practices
encourage individuals to be aware and accept their aggression-related thoughts and
emotions simply as an ephemeral state rather than to control them.
Across four studies, we investigated the possibility that brief, as opposed to
extensive, mindfulness exercise may reduce aggression, and whether this potential
effect can be separated from a general mechanism of self-control. The relationships
between mindfulness, self-control, and aggression were explored in their
dispositional forms (Study 1; N = 241). Then, the effect of brief laboratory inductions
of mindfulness was tested following manipulations designed to either bolster (Study
2; N = 99) or weaken (cross-cultural samples: Study 3; N = 119 vs. Study 4; N = 110)
the resources of self-control. In addition, the potential roles of individual differences
in sensitivity to provocations (SP) and frustrations (SF), and self-harm on aggression
were also assessed.
Results indicated that (i) despite one’s dispositional ability to exert self-control, the
presence of a mindful quality uniquely reduced the experiences of anger and
hostility, (ii) under the condition of full self-control resource (i.e., after self-control
training), mindfulness induction contributed only in reducing more subtle/implicit
forms of aggression, and (iii) under lack of self-control resource (i.e., following ego-
depleting task), mindfulness induction significantly reduced direct physical
aggression after the experience of provocation across cultures. The benefit of
mindfulness on aggression appears to be more salient when individual’s self-control
resource has been taxed, which operates similarly in Western and non-Western
settings. Therapeutic tools focusing on the mechanism for controlling the expression
of aggression would benefit from an inclusion of mindfulness-based strategies, as
well as an early identification of individual’s sensitivity to different types of
aggressive triggers and risks for self-harm.
Keywords: mindfulness induction, self-control training, ego-depletion, aggressive
behaviour, cross-cultural
iii
ACKNOWLEDGEMENT
I wish to take this opportunity to mindfully thank my supervisor Dr. Claire Lawrence.
Her knowledge, enthusiasm, and assistance have been a tremendous source of
encouragement throughout the process of the thesis.
I would also like to thank my second supervisor Dr. Ellen Townsend, my internal
examiner Dr. Peter Chapman, and my external examiner Professor Daryl O’Connor
for their thoughtful and detailed feedback at the end of year and final viva. In
addition, I am indebted to my colleagues from the Personality, Health, and Social
Psychology (PSHP) research group and from office room B76 for their academic and
social supports.
I must extend my sincerest appreciation to participants in the UK and Indonesia, who
kindly took part in the included research. This research was funded by a scholarship
from the Indonesian government, for which I am most grateful.
Special recognition goes to my family and friends for their guidance throughout. In
particular, to my flatmate Daniel Luong for all the good times we shared together.
Finally, to my twin-flame Wahyu Wicaksono: This thesis cannot be your other
without you.
iv
LIST OF CONTENTS
Abstract ii
Acknowledgement iii
List of contents iv
List of tables xi
List of figures xiii
CHAPTER ONE: Background to the thesis and literature review 1
1.1 BACKGROUND TO THE THESIS 11.1.1 Mindfulness and self-control 11.1.2 Cultural influences on mindfulness, self-control, and aggression 31.1.3 Role of individual differences 41.1.4 Aims of the thesis 4
1.2 LITERATURE REVIEW 51.2.1 Conceptualisation of aggression and aggression-related constructs 5
1.2.1.1 Definition and types of aggression 51.2.1.2 Aggression-related constructs 7
1.2.2 Mindfulness and its impact on aggression 101.2.2.1 Conceptualisation of mindfulness 101.2.2.2 Operational definition of mindfulness 101.2.2.3 Benefits of mindfulness 121.2.2.4 Mindfulness and aggression 13
1.2.3 Self control and its impact on aggression 161.2.3.1 Definition and ingredients of self-control 161.2.3.2 Strength model of self-control 181.2.3.3 Self-control and aggression 20
1.2.4 Relating mindfulness, self-control, and aggression 221.2.4.1 Distinguishing between mindfulness and self-control 221.2.4.2 Framework for mindfulness, self-control, and aggression 251.2.4.3 Role of individual differences in sensitivity to provocations and
frustrations, and self-harm 29
1.3 RESEARCH QUESTIONS AND THESIS OVERVIEW 30
CHAPTER TWO: Study 1: Relationships between trait mindfulness, self-control, andaggression 32
2.1 INTRODUCTION 322.1.1 Individual differences in mindfulness, self-control, and aggression 322.1.2 Link between mindfulness, self control, aggression, and self-harm 342.1.3 Individual differences in sensitivity to provocations and frustrations 36
v
2.2 HYPOTHESES 38
2.3 METHODS 392.3.1 Participants 392.3.2 Design 402.3.3 Measures 402.3.4 Power calculation 432.3.5 Procedure 442.3.6 Data analysis 45
2.3.6.1 Link between mindfulness, self-control, and aggression 452.3.6.2 Link between mindfulness, self-control, aggression, and self-harm
492.3.6.3 Role of individual differences in SP and SF 50
2.3.7 Data preparation 51
2.4 RESULTS 512.4.1 Influence of demographic factors 512.4.2 Trait aggression scores 522.4.3 Self-harm scores 532.4.4 Preliminary analysis 542.4.5 Mediation of self-control on the link between mindfulness and aggression
552.4.5.1 Trait aggression 552.4.5.2 Aggression subscales 56
2.4.6 Moderation of mindfulness on the link between self-control and aggression572.4.6.1 Trait aggression 582.4.6.2 Aggression subscales 58
2.4.7 Link between self-harm and harm to others 592.4.8 Link between mindfulness, self-control, and self-harm 592.4.9 Mediation of self-control on the link between mindfulness and self-harm
602.4.10 Moderation of mindfulness on the link between self-control and self-harm
612.4.11 Link between SP and SF and aggression/self-harm 612.4.12 Moderation of mindfulness and self-control on the link between SP and SF
and aggression/self-harm 612.4.12.1 Trait aggression 622.4.12.2 Aggression subscales 622.4.12.3 Self-harm 63
2.5 DISCUSSION 642.5.1 Relationships between mindfulness, self-control, and aggression 642.5.2 Link between aggression, mindfulness, self-control, and self-harm 672.5.3 Role of individual differences in SP and SF 692.5.4 Strengths and limitations of the current study 702.5.5 Conclusions and next steps 71
vi
CHAPTER THREE: Study 2: Effect of mindfulness induction and self-control training onaggressive behaviour 72
3.1 INTRODUCTION 723.1.1 Mindfulness induction and aggressive behaviour 723.1.2 Self-control training and aggressive behaviour 75
3.2 HYPOTHESES 77
3.3 METHODS 793.3.1 Participants 793.3.2 Design 793.3.3 Trait and state measures 803.3.4 Experimental manipulations 81
3.3.4.1 Self-control training 813.3.4.2 Mindfulness induction 823.3.4.3 Direct aggressive behaviour 833.3.4.5 Indirect aggression 843.3.4.6 Self-control performance 84
3.3.5 Power calculation 853.3.6 Procedure 853.3.7 Data analysis 87
3.3.7.1 Effect of mindfulness induction and self-control training onaggressive behaviour 87
3.3.7.2 Effect of mindfulness induction and self-control training on self-control performance 88
3.4 RESULTS 883.4.1 Baseline levels 883.4.2 Preliminary analysis 893.4.3 Zero-order correlations between personality variables and experimental
outcomes 903.4.3.1 Associations with direct aggression 923.4.3.2 Associations with indirect aggression 923.4.3.3 Associations with self-control performance 92
3.4.4 Manipulation Checks 933.4.5 Effect of mindfulness induction and self-control training on direct aggression
933.4.5.1 Levels of blast intensity 933.4.5.2 Maximum blast latency 973.4.5.3 Correlation between blast intensity and maximum blast latency 98
3.4.6 Effect of mindfulness induction and self-control training on indirectaggression 99
3.4.7 Correlation between blast intensity and indirect aggression 1003.4.8 Effect of mindfulness induction and self-control training on self-control
performance 101
3.5 DISCUSSION 1033.5.1 Effect of mindfulness induction and self-control training on aggressive
behaviour 103
vii
3.5.2 Effect of mindfulness induction and self-control training on self-controlperformance 106
3.5.3 Strengths and limitations of the current study 1073.5.4 Conclusions and next steps 109
CHAPTER FOUR: Study 3: Effect of mindfulness induction on aggressive behaviourfollowing depletion (British sample) 110
4.1 INTRODUCTION 1104.1.1 Self-control ego-depletion and aggressive behaviour 1104.1.2 Role of mindfulness induction in counteracting the effect of depletion on
aggressive behaviour 112
4.2 HYPOTHESES 114
4.3 METHODS 115
4.3.1 Participants 1154.3.2 Design 1164.3.3 Trait and state measures 1164.3.4 Experimental manipulations 117
4.3.4.1 Ego-depletion task 1174.3.4.2 Mindfulness induction 1184.3.4.3 Direct aggressive behaviour 1194.3.4.4 Indirect aggression 1194.3.4.5 Self-control performance 119
4.3.5 Power calculation 1204.3.6 Procedure 1204.3.7 Data analysis 122
4.3.7.1 Moderation of mindfulness induction on the link between depletionand aggressive behaviour 122
4.3.7.2 Moderation of mindfulness induction on the link between depletionand self-control performance 123
4.4 RESULTS 1244.4.1 Baseline levels 1244.4.2 Preliminary analysis 1244.4.3 Zero-order correlations between personality variables and experimental
outcomes 1254.4.3.1 Associations with direct aggression 1254.4.3.2 Associations with indirect aggression 1264.4.3.3 Associations with self-control performance 126
4.4.4 Manipulation checks 1284.4.5 Moderation of mindfulness induction on the link between depletion and
direct aggression 1294.4.5.1 Levels of blast intensity 1294.4.5.2 Maximum blast latency 1334.4.5.3 Correlation between blast intensity and maximum blast latency 134
4.4.6 Moderation of mindfulness induction on the link between depletion andindirect aggression 135
viii
4.4.7 Correlation between blast intensity and indirect aggression 1374.4.8 Moderation of mindfulness induction on the link between depletion and
self-control performance 138
4.5 DISCUSSION 1394.5.1 Effect of mindfulness induction on aggressive behaviour following depletion
1394.5.2 Effect of mindfulness induction on self-control performance following
depletion 1424.5.3 Strengths and limitations of the current study 1424.5.4 Conclusions and next step 144
CHAPTER FIVE: Study 4: Effect of mindfulness induction on aggressive behaviour followingdepletion (Indonesian sample) 145
5.1 INTRODUCTION 1455.1.1 Mindfulness and Indonesian culture 1455.1.2 Indonesian culture and aggression 1465.1.3 Indonesian culture and self-control 1485.1.4 Role of mindfulness induction in counteracting the effect of depletion on
aggressive behaviour for Indonesian culture 148
5.2 HYPOTHESES 149
5.3 METHODS 1495.3.1 Participants 1495.3.2 Design 1505.3.3 Trait and state measures 1505.3.4 Experimental manipulations 1515.3.5 Power calculation 1515.3.6 Procedure 1515.3.7 Data analysis 152
5.4 RESULTS 1525.4.1 Baseline levels 1535.4.2 Preliminary analysis 1535.4.3 Correlation between personality variables and experimental outcomes 154
5.4.3.1 Associations with direct aggression 1545.4.3.2 Associations with indirect aggression 1555.4.3.3 Associations with self-control performance 155
5.4.4 Manipulation Checks 1575.4.5 Moderation of mindfulness induction on the link between depletion and
direct aggression 1585.4.5.1 Levels of blast intensity 1585.4.5.2 Maximum blast latency 1615.4.5.3 Correlation between blast intensity and maximum blast latency 163
5.4.6 Moderation of mindfulness induction on the link between depletion andindirect aggression 163
5.4.7 Correlation between blast intensity and indirect aggression 165
ix
5.4.8 Moderation of mindfulness induction on the link between depletion andself-control performance 165
5.5 DISCUSSION 1675.5.1 Effect of mindfulness induction on aggressive behaviour following depletion
1675.5.2 Effect of mindfulness induction on self-control performance following
depletion 1695.5.3 Strengths and limitations of the current study 1705.5.4 Conclusions and next steps 171
CHAPTER SIX: Cross-cultural comparison of the effect of mindfulness induction onaggressive behaviour following depletion (British vs. Indonesian sample) 172
6.1 INTRODUCTION 1726.1.1 Cultural differences in the effect of mindfulness induction on aggressive
behaviour following depletion 1726.1.2 Cultural differences in the effect of mindfulness induction on self-control
performance following depletion 1746.1.3 Cultural differences in personality variables 175
6.2 HYPOTHESES 176
6.3 METHODS 1786.3.1 Participants 1786.3.2 Data analysis 179
6.4 RESULTS 1806.4.1 Cultural differences in personality variables 1806.4.2 Cultural differences in the moderation of mindfulness induction on the link
between depletion and aggressive behaviour 1816.4.2.1 Levels of blast intensity 1816.4.2.2 Maximum blasts latency 1846.4.2.3 Indirect aggression 185
6.4.3 Cultural differences in the moderation of mindfulness induction on the linkbetween depletion and self-control performance 187
6.5 DISCUSSION 1886.5.1 Cultural differences in the effect of mindfulness induction on aggressive
behaviour following depletion 1886.5.2 Cultural differences in the effect of mindfulness induction on self-control
performance following depletion 1906.5.3 Cultural differences in personality variables 1916.5.4 Cultural differences in aggressive behaviour 1926.5.5 Cultural differences in self-control performance 1946.5.6 Strengths and limitations of the current analysis 1956.5.7 Conclusions 196
x
CHAPTER SEVEN: Summary and general discussion 197
7.1. SUMMARY 1977.1.1 Summary of each study 1977.1.2 Summary of findings 200
7.2. GENERAL DISCUSSION 2017.2.1 Mindfulness, self-control, and aggression 2017.2.2 Cultural differences in sensitivity to provocation and aggressive triggers 2057.2.3 Mindfulness, self-control, and self-harm 207
7.3 LIMITATIONS AND IMPLICATIONS FOR FUTURE RESEARCH 2087.3.1 Sample characteristics 2087.3.2 Brief laboratory mindfulness induction 2097.3.3 Self-control task 2117.3.4 Aggressive triggers 2127.3.5 Self-reported measures 2147.3.6 Future directions 215
REFERENCES 219
APPENDICES 239
Appendix 2.1: Trait measures for Study 1 to 4 239Appendix 2.2: Advertising materials for Study 1 247Appendix 2.3: Information to participants for Study 1 248Appendix 2.4: Electronic consent form for Study 1 and Study 2 (Time 1) 249Appendix 2.5: Debrief materials for Study 1 250
Appendix 3.1: State measures for Study 2, 3 and 4 251Appendix 3.2: Daily remainder for self-control group in Study 2 253Appendix 3.3: Mindfulness instruction for Study 2 254Appendix 3.4: Relaxation instruction for Study 2 255Appendix 3.5: Advertising materials for Study 2 257Appendix 3.6: Information to participants for Study 2 (Time 1) 258Appendix 3.7: Debrief materials for Study 2 (Time 1) 259Appendix 3.8: Information to participants for Study 2 (Time 2), and Study 3 and 4
260Appendix 3.9: Consent form for Study 2 (Time 2), and Study 3 and 4 261Appendix 3.10: Debrief materials for Study 2 (Time 2), and Study 3 and 4 262
Appendix 4.1: Mindfulness instruction for Study 3 and 4 263Appendix 4.2: Advertising materials for Study 3 and 4 (trans.) 265
Appendix 6.1: Summary of findings for cross-cultural analysis 266
xi
LIST OF TABLES
Table 2.1: Means for aggression measures based on participants’ sex 52Table 2.2: Zero order correlations and psychometric properties of measures 54Table 2.3: Mediation models with bootstrapping method, using mindfulness as
predictor, self-control as mediator, aggression subscales as outcomes 56Table 2.4: Partial effect of control variables for mediation models with bootstrapping
method, using mindfulness as predictor, self-control as mediator, aggressionsubscales as outcomes 57
Table 2.5: Moderation models with bootstrapping method, using self-control aspredictor, mindfulness as moderator, aggression subscales as outcomes 58
Table 2.6: Moderation models with bootstrapping method, using sensitivity toprovocations (SP) as predictor, mindfulness and self-control as moderators,and aggression subscales as outcomes 63
Table 2.7: Moderation models with bootstrapping method, using sensitivity tofrustrations (SF) as predictor, mindfulness and self-control as moderators,and aggression subscales as outcomes 63
Table 3.1: Zero order correlations and psychometric properties of measures 89Table 3.2: Zero order correlations of personality measures, aggressive behaviour, and
self-control performance 91Table 3.3: Sex differences on blast intensity on each provocation level 94Table 3.4: Moderation models with bootstrapping method, using self-control training
as predictor, mindfulness as moderator, blast intensities as outcome 96Table 3.5: Moderation models with bootstrapping method, using self-control training
as predictor, mindfulness as mediator, blast intensity as outcome,controlling for covariates 96
Table 3.6: Moderation models with bootstrapping method, using self-control trainingas predictor, mindfulness as moderator, rating of opponent as outcome 99
Table 3.7: Sex differences based on rating of opponent 99Table 3.8: Moderation models with bootstrapping method, using self-control training
as predictor, mindfulness as moderator, rating of opponent as outcome,controlling for covariates 100
Table 3.9: Bivariate correlations between blast intensity and rating of opponent 100Table 3.10: Moderation models with bootstrapping method, using self-control training
as predictor, mindfulness as moderator, handgrip power as outcome 102
Table 4.1: Zero order correlations and psychometric properties of measures 124Table 4.2: Zero order correlations of personality measures, aggressive behaviour, and
self-control performance 127Table 4.3: Sex differences on blast intensity on each provocation level 130Table 4.4: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, blast intensity as outcome 131Table 4.5: Partial effects of control variables for moderation models with
bootstrapping method, using depletion as predictor, mindfulness asmoderator, blast intensity as outcome 132
Table 4.6: Moderation models with bootstrapping method, using depletion aspredictor, mindfulness as moderator, rating of opponent as outcome 135
Table 4.7: Sex differences based on rating of opponent 136
xii
Table 4.8: Moderation models with bootstrapping method, using depletion aspredictor, mindfulness as moderator, rating of opponent as outcome,controlling for covariates 136
Table 4.9: Bivariate correlations between blast intensity and rating of opponent 137Table 4.10: Correlations between blast intensity and rating of opponent based on sex
137Table 4.11: Overall change in handgrip performance 138Table 4.12: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, changes in handgrip performance asoutcome, controlling for covariates 138
Table 5.1: Zero order correlations and psychometric properties of measures 153Table 5.2: Zero order correlations of personality measures, aggressive behaviour, and
self-control performance 156Table 5.3: Sex differences on blast intensity on each provocation level 159Table 5.4: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, blast intensity as outcome 159Table 5.5: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, blast intensities as outcome,controlling for covariates 161
Table 5.6: Moderation models with bootstrapping method, using depletion aspredictor, mindfulness as moderator, rating of opponent as outcome 164
Table 5.7: Sex differences based on rating of opponent 164Table 5.8: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, rating of opponent as outcome,controlling for covariates 165
Table 5.9: Overall change in handgrip performance 166Table 5.10: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness as moderator, changes in handgrip performance asoutcome, controlling for covariates 166
Table 6.1: Frequency distributions for cross-cultural analysis 178Table 6.2: Cross-cultural differences in personality measures 180Table 6.3: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness and culture as moderator, blast intensity as outcome182
Table 6.4: Means for blast intensity across culture and sex 182Table 6.5: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness and culture as moderator, rating of opponent asoutcome 185
Table 6.6: Means for rating of opponent across culture and sex 186Table 6.7: Moderation models with bootstrapping method, using depletion as
predictor, mindfulness and culture as moderator, changes in handgripperformance as outcome 187
Table 6.8: Overall change in handgrip performance across culture and sex 188
xiii
LIST OF FIGURES
Figure 2.1: Simple mediation model 46Figure 2.2: Conceptual and statistical models for simple moderation 48Figure 2.3: Conceptual and statistical models for additive moderation 50
Figure 3.1: Flow of participants through Study 2 80Figure 3.2: Effect of level of provocations on blast intensity 94Figure 3.3: Mindfulness x self-control training effect on blast intensity 95Figure 3.4: Sex differences in maximum blast latency 98Figure 3.5: Crossover interaction between mindfulness x self-control training on
handgrip power 101Figure 3.6: Sex differences in handgrip power 102
Figure 4.1: Flow of participants through Study 3 116Figure 4.2: Effect of level of provocations on blast intensity 129Figure 4.3: Mindfulness × depletion effect on blast intensity across provocation levels
131Figure 4.4: Sex differences in maximum blast latency 134Figure 4.5: Sex differences in handgrip stamina 139
Figure 5.1: Flow of participants through Study 4 150Figure 5.2: Effect of level of provocations on blast intensity 158Figure 5.3: Mindfulness × depletion effect on blast intensity across provocation levels
160Figure 5.4: Sex differences in maximum blast latency 162Figure 5.5: Sex differences in handgrip stamina 167
Figure 6.1: Conceptual and statistical models for moderated moderation 179Figure 6.2: Cross-cultural comparisons of blast intensities based on experimental
conditions 183
CHAPTER ONE
Background to the thesis and literature review
1.1 BACKGROUND TO THE THESIS
Aggressive behaviour can be problematic in a wide array of settings, including in
school, workplace, mental health rehabilitation, and correctional facility. The latest
figures from the Crime Survey for England and Wales (CSEW) estimate 2.1 million
violent incidents in England and Wales in 2011/12, with 3% of adults victimised
(Office for National Statistics, 2013). The CSEW figures are based on a representative
sample survey of adults resident in households in England and Wales about the
crimes they have experienced in the previous 12 months, including crimes that have
not been reported to or recorded by the police.
In Western countries, various psychological interventions have been employed to
predict and prevent aggression and violence. Throughout the past two decades, the
majority of psychological treatments for offender populations were based on
“second-wave” cognitive-behavioural approaches (Howells, Tennant, Day, & Elmer,
2010). Within these approaches, individuals are taught various strategies that enable
them to make a deliberate attempt to control and modify their undesirable thoughts
or emotions that may produce maladaptive behaviours.
1.1.1 Mindfulness and self-control
More recently, a group of meditational practices known as mindfulness techniques
has been promoted as part of the application of “third-wave” cognitive-behavioural
approaches for aggression (Howells, 2010; Wright, Day, & Howells, 2009). Third-
wave approaches are strongly influenced by Eastern contemplative traditions,
Buddhism in particular. Essentially, the goal of these approaches is to encourage
individuals to observe and accept, rather than to exert control over their thoughts,
2
emotions, or behaviours (a thorough discussion on mindfulness, acceptance, and
cognitive-behavioural tradition is provided in S. C. Hayes, Follett, & Linehan, 2004).
While debate over the precise definition of mindfulness continues, Western
researchers commonly follow Bishop et al.’s (2004) operationalisation of mindfulness
as two components of (i) sustained attention on the immediate experience, and (ii)
an attitude of curiosity, openness, and acceptance towards one's experiences.
Sustaining attention on uncomfortable experiences, without making any direct
attempt to modify (e.g., avoid, suppress, ruminate on) them, is thought to increase
one’s exposure to these experiences, and should eventually lead to the extinction of
any habitual associated responses (e.g., Baer, 2003; Linehan, Bohus, & Lynch, 2007;
Shapiro, Carlson, Astin & Freedman, 2006). Arguably, mindfulness may reduce
individual’s emotional reactivity that is typically associated with aggression-
triggering events.
Increasingly, the current literature has documented the application of mindfulness-
based interventions for treatment of aggression in mental health and forensic
settings. Despite encouraging evidence, significant methodological flaws, including
lack of randomised-controlled trials, sole reliance on self-reported measures, and
inconsistent definitions of mindfulness, have been identified (see Fix & Fix, 2013;
Shonin, Gordon, Slade, & Griffiths, in press). Crucially, Masicampo and Baumeister
(2007) have pointed out many similarities between features of mindfulness-based
interventions and training in self-control. They maintain that the very reason for the
various benefits of mindfulness interventions (e.g., physical health, mental health,
behaviour regulation, emotion regulation, interpersonal relationships) because
mindfulness training is, essentially, an example of self-control training. As a result,
the distinct contribution of mindfulness per se has not been established.
The ability to control oneself is relevant to aggression because the inability to refrain
from behaving aggressively is typically an undesirable response in modern societies.
Indeed, an accumulating body of research has documented the link between self-
control and aggression (Caspi, 2000; Gottfredson & Hirschi, 1990; Moffitt et al.,
2011). If the benefits of mindfulness on aggression is entirely attributable to a
3
general mechanism of self-control, then being more mindful when encountering
aggression-triggering events should not produce any additional reduction in
aggression. However, as we will describe in the next section, mindfulness and self-
control appear to be conceptually different. Consequently, it is feasible that the
extent to which each approach may influence aggression may vary. To our
knowledge, no published studies to date have particularly focus on any differential
association between mindfulness, self-control, and aggression.
It is a key aim of the thesis, therefore, to identify whether and how the effect of
mindfulness on aggression can be disentangled from the effect of self-control,
particularly outside extensive training in mindfulness. For this purpose, we shall
focus on aggression in general instead of the specific forms of aggression such as
violence or sexual offending.
1.1.2 Cultural influences on mindfulness, self-control, and aggression
Cultures determine their own sets of values to which individuals are exposed.
Individuals from different cultures may vary in their familiarity and acceptance of
mindfulness and the application of mindfulness in daily life; whereas self-controlled
regulation of personal goals and professional achievement is pervasive in Western
cultures (Brown, Ryan, & Creswell, 2007a). In addition, tolerance towards different
types of aggression may also vary across cultures. Previous cross-cultural studies
(e.g., Bergeron & Schneider, 2005; Forbes, Zhang, Doroszewicz, & Haas, 2009;
French, Jansen, & Pidada, 2002) have shown that individuals from a culture that
emphasises maintaining social harmony (typically associated with Eastern and
collectivist cultures) are less likely to resort to direct methods of aggression in
response to provocation than those from individualistic cultures. These cultural
characteristics may also moderate any influence of mindfulness, and self-control on
aggressive behaviour. Examining the utility of the Western operationalisation of
mindfulness amongst individuals from Eastern cultures is also important, considering
the cross-cultural validation of this Western operationalisation is still lacking
(Christopher, Charoensuk, Gilbert, Neary, & Pearce, 2009).
4
1.1.3 Role of individual differences
Individuals may vary not only in terms of their general propensity to aggress, but also
in their sensitivity to different aggressive triggers. Lawrence (2006) have shown that
individual differences in sensitivity to provocations (SP) and frustrations (SF) were
associated to distinct components of aggression. Thus it is plausible that SP and SF
may have different influence on the potential link between mindfulness, self-control,
and aggression. To date, this possibility has not been examined in the literature.
It is also noteworthy that the potential link between mindfulness, self-control, and
aggression is investigated in relation to individuals’ tendency to harm themselves
(self-harm). The literatures on aggression and self-harm are commonly separated.
However, similar possible antecedents of these two harmful behaviours have been
identified (e.g., Hillbrand, 2001; Placidi et al., 2001; Plutchick & van Praag, 1989;
Roaldset, Bakken, & Bjørkly, 2011). This highlights the necessity for setting up more
integrated risk assessments of aggression and self-harm. Even though mindfulness-
based interventions have been widely applied for treatment of self-harming
behaviour (Linehan, 1993; Segal, Williams, & Teasdale, 2012), the potential and
relative benefits of mindfulness and self-control on self-harm have rarely been
investigated in a single study (for an overview, see Slee, 2008).
1.1.4 Aims of the thesis
The main purposes of this thesis are to investigate the role of mindfulness on
aggression, and to elaborate whether and when this potential effect can be
separated from a general mechanism of self-control. In particular, the thesis aims to
do the following:
1. Examine the extent to which the benefits of mindfulness on aggression may
occur outside extensive training in mindfulness.
2. Directly compare the benefits of mindfulness and self-control on aggression.
3. Assess the validation of Western operationalisation of mindfulness, and self-
control, on aggression.
5
4. Investigate any similarities between the mechanisms of aggression and self-
harm regarding mindfulness and self-control and individual differences.
As a whole, the thesis aims to provide a useful framework on whether and when
mindfulness and self-control could be incorporated in the intervention strategies for
the reduction of aggression and self-harm. The information that can be obtained
from this investigation may potentially contribute both clinically and practically.
1.2 LITERATURE REVIEW
In this section, we will first outline the conceptualisation of aggression and
aggression-related constructs. Then we will review the potential role of mindfulness
in the reduction of aggression. Our focus will then turn to current evidences for the
effect of self-control on aggression, with a particular focus on the strength model of
self-control. Finally, we will suggest a framework to support our proposal that
mindfulness may influence aggression in the same way it is proposed to be
influenced by self-control, but that the effect of mindfulness may even be
independent of and beyond the effect of self-control.
1.2.1 Conceptualisation of aggression and aggression-related constructs
1.2.1.1 Definition and types of aggression
Human aggression has been traditionally dichotomised into (i) hostile aggression,
which refers to impulsive, reactive behaviour driven by anger, motivated by a desire
of harming the target, and (ii) instrumental aggression, which refers to
premeditated, proactive behaviour motivated by goals other than harming the target
(e.g., B. A. Baron & Richardson, 1994; Berkowitz, 1993; Geen, 2001). However, this
view has failed to consider that aggressive acts may be motivated by many different
goals (for a comprehensive argument, see Bushman & Anderson, 2001).
6
To allow a discussion between hostile and instrumental aggression, Bushman and
Anderson (2001) define aggression as any behaviours directed towards another
individual with the proximate/immediate intent to cause harm to the target,
whereas the target is motivated to avoid the behaviour (a similar definition is
provided in B. A. Baron & Richardson 1994; Berkowitz, 1993; Geen, 2001). While the
presence of intention to harm others is necessary as a proximate goal, different acts
of aggression may vary in terms of their ultimate goals. For instance, robbery and
physical assault are classified as aggression since both conducts include intention to
harm the target at a proximate level; although ultimately robbery also serves profit-
based goals and assault serves harm-based goals (Anderson & Bushman, 2002).
Bushman and Anderson’s definition of aggression can be clearly seen when the
perpetrator directly inflicts harm on a target, either physically (e.g., hitting another
person) or verbally (e.g., yelling at someone), in a face-to-face situation or where the
perpetrator is easily identified. Human beings, however, are capable of inflicting
harm through concealed acts, such as damaging the reputation, gossiping, or socially
excluding others. This type of aggression is defined by Björkqvist and colleagues
(Björkqvist, 1994; Björkqvist, Lagerspetz, & Kaukianen, 1992) as indirect aggression;
due to its circuitous and anonymous delivery of harm. Much like direct aggression,
indirect aggression can be either physical (e.g., hiring an assassin) or verbal (e.g.,
spreading rumour). In an integrated review, Archer and Coyne (2005) conclude that
indirect aggression is essentially the same form of aggression with other constructs
labelled as relational aggression (Crick & Grotpeter, 1995) and social aggression
(Cairns, Cairns, Neckerman, Ferguson, & Gariepy, 1989), in which the perpetrator’s
intention is to socially exclude, or harm the social status of the target.
This thesis employs the definition of aggression as a behavioural process of harming
the target who is motivated to avoid the act, and may be carried out through a direct
or indirect route, either physically or verbally. Thus aggression excludes behaviour in
the absence of any intention to harm (e.g., accidentally stepping on someone’s foot),
behaviour where the target is not motivated to avoid (e.g., suicidal behaviour, sexual
masochism), as well as prosocial acts that happen to cause harm to the target (e.g., a
surgeon causing pain to her patient during surgery).
7
1.2.1.2 Aggression-related constructs
The criterion problem in aggression research often occurs since the terms aggressive
behaviour, anger, and hostility have been used interchangeably (Parrot & Giancola,
2007). Although anger is a common antecedent of aggression, it does not always
lead to aggressive behaviour. Most individuals report experiencing mild to moderate
levels of anger from several times per day to several times per week, but they
usually do not act upon it (Averill, 1983). As defined by Berkowitz (1993), anger is an
emotional response to an event or situation that lacks of a specific behavioural goal,
whereas hostility is a cognitive construct that involves negative interpretations and
thoughts about the environment.
In a similar vein, Buss and Perry (1992) define (i) anger as the emotional or affective
component of aggression that involves physiological arousal and preparation for
aggression, and (ii) hostility as the cognitive component of aggression which consists
of feelings of ill-will and injustice. They suggest that because anger is a high-arousal
state that diminishes over time, dissipation of anger may cause a cognitive residual
of hostility. Both anger and hostility, in combination with physical and verbal
aggression which represent the instrumental, motor, or overt component of
behaviour, are subsumed under the general trait of aggression (Buss & Perry, 1992).
D. B. O’Connor, Archer, and Wu (2001) demonstrated that trait aggression was
related to aggressive response to provoking scenario (as measured using a vignette-
based approach), and that aggressive response was related to aggressive feelings
(i.e., anger, frustration, and irritation). This indicates that trait aggression may set up
a basis for individuals’ preparedness to behave aggressively when encountering
situational triggers of aggression. As shown in the General Aggression Model (GAM:
Anderson & Bushman, 2002), the person‐based (who) and situation-based (when)
predictors of aggression would influence the production of an aggressive or
nonaggressive behaviour (why) through their influence on the present internal states
(i.e., cognition, affect, and arousal routes) that determine the underlying appraisal
and decision processes.
8
Lawrence and colleagues (Lawrence, 2006; Lawrence & Hodgkins, 2009; Lawrence &
Hutchinson 2013b) have further developed a distinction between provocations (i.e.,
goading and instigation from others) and frustrations (i.e., situations where the
individual experiences a lack of control due to circumstances or the behaviour of
others) as two independent triggers of aggression. Their initial study showed that
individuals vary in the extent to which they were sensitive to these two aggressive
triggers; and that sensitivity to provocations (SP) was related to self-reported overt
aggression, whereas sensitivity to frustrations (SF) was related to hostility and anger
(Lawrence, 2006). Individuals’ differences in SP, but not in SF, also predicted
aggressive interpretations of provoking individuals beyond trait aggression and
mood (Lawrence & Hodgkins, 2009) and aggressive behaviour under provoking
situation after accounting for trait aggression and participants’ sex (Lawrence &
Hutchinson, 2013b).
Individual differences in SP and SF, as well as in the general propensity to aggress,
may differentially influence the extent to which individuals are more likely to behave
aggressively under different aggressive triggers. As this thesis examines aggressive
behaviour in response to provoked and unprovoked aggression, it is necessary to
account for the potential influences of SP, SF, and trait aggression on the link
between mindfulness, self-control, and aggression.
Further, the current literature has also documented similar antecedents for
aggression and self-harm, including lower levels of serotonin (Barbui et al., 2009;
Roaldset et al., 2011), cerebrospinal fluid monoamine metabolite (Placidi et al.,
2001), greater experience of hopelessness and higher levels of impulsivity (Plutchick
& van Praag, 1989), and severe lack of behavioural control (Hilbrand, 2001). This
indicates that individuals’ propensity to cause harm to others may also be related to
their tendency to harm themselves.
Gratz (2001) defines “deliberate self-harm” as the deliberate, direct destruction or
alteration of body tissue without conscious suicidal intent, but resulting in injury
severe enough for tissue damage (e.g., scarring) to occur. A similar definition is
referred by Favazza (1996, 2012) and Nock (2010) as non-suicidal self-injury. With
9
this definition, a distinction is made between self-harming behaviours that are
suicidal and non-suicidal in nature. In this thesis, we ask participants to complete
Gratz’s (2001) measure of deliberate self-harm, which was designed to specifically
assess self-harm in the absence of suicidal intent.
More recently, however, the prefix “deliberate” is no longer preferred by the
National Collaborating Centre for Mental Health (NCCMH, 2012) because it is
considered judgemental, and the intentional aspect of self-harm is not always clear.
Indeed, the presence/absence of suicidal intent may be unclear even to the
individuals who harm themselves; and repetition of self-harm, regardless of the
intention, is a risk factor for suicide (see Kapur, Cooper, R. C. O'Connor, & Hawton,
2013). Following the NCCCMH’s recommendation, we use the term self-harm to
allow a discussion about this behaviour regardless the underlying motivation. Self-
harm also connotes other similar constructs such as self-injury, self-mutilation, self-
poisoning, and self-aggression that have been used interchangeably in the literature.
The examination of the relationship between mindfulness, self-control, aggression
and self-harm in this thesis is conducted, in order to explore the extent to which
mindfulness and self-control influence self-harm in the same way as they influence
aggression. Self-harm is also examined due to its potential influence on the link
between mindfulness, self-control, and aggression.
In the following section, the original conceptualisation and recent Western
operationalisation of mindfulness will be presented. Then the current empirical
findings for the benefits of mindfulness on various psychological and physical
functioning will be highlighted, followed by preliminary evidences for potential role
of mindfulness in the reduction of aggression.
10
1.2.2 Mindfulness and its impact on aggression
1.2.2.1 Conceptualisation of mindfulness
The concept of mindfulness originates from Eastern contemplative practices,
specifically the Buddhist tradition of mindfulness meditation. The term
“mindfulness” is translated from the Pali word (the language in which the early
Buddhist teaching was recorded), sati, which literally means “memory”, and closely
linked to the verb, sarati, which means “to remember”. The word sati also connotes
attention (Shapiro & Carlson, 2009), and relates frequently with another Pali word,
sampajanna, which means awareness or clear comprehension (Dryden & Steele,
2006; Vago & Silbersweig, 2012). Thus mindfulness does not refer to remembering
to memory of past events, but to remembering to pay attention to and be aware of
all that is taking place in the present moment (e.g., Siegel, Germer, & Olendzki, 2009;
Wallace & Bodhi, 2006).
Over the last two decades, Western mental health professionals and researchers
have extensively incorporated mindfulness meditation independent of any religious
system, mainly as a therapeutic intervention technique. The mental qualities beyond
sati, such as nonjudgment and acceptance, are typically included in the
contemporary conceptualisation of mindfulness (Bishop et al., 2004; Siegel et al.,
2009). This corroborates with one of the most often cited Western definitions of
mindfulness as awareness that arises through “paying attention in a particular way:
on purpose, in the present moment, and nonjudgementally” (Kabat-Zinn, 1994, p.4).
1.2.2.2 Operational definition of mindfulness
Given the richness of the tradition from where mindfulness originates, Western
researchers have continuously worked to develop an empirically-based operational
definition of mindfulness, primarily validated through means of self-reported
measures. The prevailing view appears to follow Bishop et al.’s (2004) definition of
two-component model, in which mindfulness consists of (i) self-regulation of
attention, and (ii) adoption of a particular orientation towards one's experiences.
11
Self-regulation of attention refers to nonelaborative awareness of thoughts,
emotions, or body sensations as they arise in the present moment. Orientation to
experience is characterised by attitude of curiosity, openness, and acceptance.
Acceptance in this context is not passivity or resignation, but an active process of
nonjudgmental exposure to the “in-the-moment” experience (Cardaciotto, Herbert,
Forman, Moitra, & Farrow, 2008; S. C. Hayes, 2004).
Bishop and colleagues asserted that mindfulness is a mode of awareness cultivated
when attention is regulated in the manner described, emphasising its state-like
quality. State of mindfulness can be observed immediately through brief mindfulness
interventions (Brown & Cordon, 2009), and may be converted into a stable tendency
if practiced over a period of time (Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006;
Chambers, Gullone, & Allen, 2009).
On the other hand, the most prominently used mindfulness scale to date, the
Mindful Awareness Attention Scale (MAAS) is based on Brown and Ryan’s (2003)
operationalisation of mindfulness as a unidimensional construct of receptive
attention to and awareness of on-going events and experience in daily life. An
accumulative body of research has found a great variation of the MAAS score
amongst non-meditating and untrained individuals, underlining an inherent human
capacity of mindfulness (Brown & Ryan, 2003; Brown, Ryan, Loverich, Biegel, &
West, 2011; see also Kabat-Zinn, 2003). When described as a trait, the “acceptance”
quality of mindfulness is seen as a prerequisite element for being fully engaged with
the present experience (an extensive argument is provided in Brown & Ryan, 2004).
Although currently speculative, Holzel et al. (2011) and Sauer et al. (2013) suggest
that whether acceptance is an independent component or a prerequisite for
mindfulness may be attributable to the degree of meditation expertise. Specifically,
in the normal states of mind, emotional reactions occur automatically. To be able to
sustain attention on the present moment, a novice practitioner might need to exert
cognitive control over the emotional reactions. With increasing mindfulness practice,
one may have automated an accepting attitude, thus attention can be sustained in
12
an effortless way (for a recent review on the functional neural mechanisms of
mindfulness, see Chiesa, Serretti, & Jakobsen, 2013).
For the purpose of this thesis, we follow Brown and Ryan’s conceptualisation when
describing mindfulness as a relatively stable individual dispositional, and assess the
Bishop et al.’s components of mindfulness when examining the state aspect of
mindfulness. While these authors’ conceptualisations are not identical, they both
imply that sustained attention represents the element most central to mindfulness.
1.2.2.3 Benefits of mindfulness
Empirical research on mindfulness has been conducted through (i) clinical
interventions, to examine the effect of mindfulness training on psychological and
physical functioning, (ii) correlational studies, to examine the relations between
individual differences in mindfulness and other constructs, and (iii) laboratory study,
to test the immediate effects of brief mindfulness inductions. Although attentional
control in emotion regulation literature is typically seen as adaptive when attention
is directed away from emotionally provocative stimuli, mindfulness encourages
individuals to bring attention back to the stimulus (Holzel et al., 2011).
A number of mindfulness-based clinical interventions, namely Mindfulness-Based
Stress Reduction (MBSR: Kabat-Zinn, 1990), Mindfulness-Based Cognitive Therapy
(MBCT: Segal et al., 2012), Acceptance and Commitment Therapy (ACT: S. C. Hayes,
Strosahl, & Wilson, 2011), Dialectical Behaviour Therapy (DBT: Linehan, 1993), and
Relapse Prevention (RP: Bowen, Chawla, & Marlatt, 2010) have been applied in
treating a broad range disorders, including pain, stress, anxiety, depressive relapse,
and eating disorders, as well as in improving overall mental health and well-being.
Since 2009, the MBCT has been recommended by the National Institute for Health
and Clinical Excellence (NICE) in the UK as an effective programme for preventing
depressive relapse. Indeed, the strongest clinical effect sizes of mindfulness have
been reported for the direct treatment of anxiety and/or mood-spectrum disorders
(Shonin, Gordon, & Griffiths, 2013). Even eight weeks of mindfulness practice may
13
change the way emotion is processed in the brain (Williams, 2010). However,
mindfulness-based interventions have been criticised for not consistently examining
the relative contribution of mindfulness independently of other behaviour change
strategies (for a review, see e.g., Baer, 2003; Chiesa & Malinowski, 2011; Davis & J.
A. Hayes, 2011).
The inclusion of self-reported measures and laboratory studies of mindfulness may
allow researcher to scrutinise the nature of mindfulness through its associations and
influences on other related variables. Accordingly, in a current review on the effect
of mindfulness, Keng, Smoski, and Robins (2011) have outlined associations between
trait mindfulness with a host of indicators of psychological health, such as higher
levels of positive affect, sense of autonomy, vitality, and adaptive emotion
regulation, and lower levels of negative affect, rumination, cognitive reactivity, and
general psychological symptoms, as well as differences in brain activity observed
using functional neuroimaging methods.
Keng et al. have also concluded the effect of brief laboratory mindfulness inductions
on various emotion-related processes, such as recovery from dysphoric mood,
emotional reactivity to aversive or emotionally provocative stimuli, and willingness
to return to or persist on an unpleasant task. These findings suggest that brief
mindfulness inductions may work in a similar way as mindfulness training given over
a longer period. Hence, some benefits of mindfulness may possibly be measured
immediately in the absence of extensive prior training in mindfulness.
1.2.2.4 Mindfulness and aggression
As a result of the benefits of mindfulness on reducing negative emotions, it is
reasonable to expect mindfulness to attenuate the associations between aggressive
triggers and aggression-related emotions, such as anger. Since aggressive emotions,
thoughts, and behavioural tendencies are linked together in memory (see cognitive
neoassociation theories, e.g., Berkowitz, 1989, 1990, 1993), mindfulness may even
limit the activation of this link in the first place.
14
Current applications of mindfulness in the area of aggression have focused on the
effects of mindfulness-based interventions in mental health and forensic settings.
According to Howells et al. (2010), the incorporation of mindfulness into pre-existing
cognitive-behavioural frameworks is particularly advantageous in situations when
the individual reacts too quickly to consciously change a thought or behaviour, when
emotional responses overwhelm conscious control of behaviour, or when the
individual is lack of introspective ability and self-awareness to apply the conventional
cognitive-behavioural strategies. As such, the risk factors of aggression that
potentially curable through mindfulness also include poor emotional self-regulation
(e.g., impulsivity), in addition to problems of negative emotions (e.g., anger).
In correctional facility settings, mindfulness-based interventions have been reported
to reduce self-reported hostility (Samuelson, Carmody, Kabat-Zinn, & Bratt, 2007), as
well as physical interventions, seclusions, and incidences of self-aggression (Chilvers,
Thomas, & Stanbury, 2011). In a series of case-study, Singh and colleagues
demonstrated the efficacy of a mindfulness-based intervention as a self-
management technique to control aggressive behaviour in individuals with mild
intellectual deficit and mental illness (2003), conduct disorder (2007), and autism
(2011). Specifically, these individuals were taught to focus their attention mindfully
on the “soles of the feet” when encountering aggression-provoking thoughts, events,
or situations, then to make a choice about how to react once the calmness and
clarity of mind have been established. A randomised controlled trial effect of the
Soles of the Feet technique was recently reported for reducing verbal and physical
aggression in individuals with mild intelectual disability (Singh et al., 2013).
Correlational studies of mindfulness have also supported associations between
individuals’ tendency to be mindful and their propensity to aggress. In the
development study of the MAAS, Brown and Ryan (2003) found that trait
mindfulness was negatively related to self-reported anger and hostility amongst
breast and prostate cancer patients. Studies in samples of general population
(Borders, Earleywine & Jajodia, 2010) and undergraduates (Kelly & Lambert, 2012)
also found negative associations between trait mindfulness and self-reported anger,
hostility, verbal aggression, and physical aggression. In addition to reporting lower
15
general disposition of aggressiveness, mindful individuals interpreted ambiguous
social information as being less hostile (Heppner et al., 2008; Kelly & Lambert, 2012).
An exploratory attempt to investigate the effect of mindfulness in laboratory setting
was conducted by Barnes, Brown, Krusemark, Campbell, and Rogge (2007). In their
study, sixty heterosexual couples were asked to discuss their about relationship
conflict topics, and their cognitive, emotional, and behavioural responses were
examined. Following a conflict discussion, individuals with higher levels of trait
mindfulness reported less anger-hostility towards their romantic partners. This
relationship was mediated by individual’s level of anger-hostility before the conflict
discussion. Moreover, higher state mindfulness, but not trait mindfulness, predicted
a better video-coded communication quality, including less verbal aggression (i.e.,
hostile and aggressive remarks directed towards the partner) and negativity and
conflict (i.e., display of tension, irritation, anger), and marginally predicted less
withdrawal (i.e., avoiding the interaction or discussion).
It should be noted, however, that Barnes et al.’s study did not actively induce
participants’ level of mindfulness. Instead, participants were asked to rate how
mindful they were during the induction of conflict. Still, their study revealed that the
potential effect of mindfulness on expressed aggression, in addition to its more
typical effect on aggression-related emotions (i.e., anger) and cognitions (i.e.,
hostility), may occur outside the context of extensive training in mindfulness.
To our knowledge, the only reported aggression study to date that actively
manipulated participants’ state of mindfulness was conducted by Heppner et al.
(2008). Specifically, a total of sixty undergraduates were randomly assigned to one of
three conditions in a social rejection task: acceptance from partner, rejection by
partner, or mindfulness plus rejection. A state of mindfulness was induced before
the rejection feedback, using a 5-min task involving eating raisins while being
focused in a mindful way on the experience of eating the raisin from Kabat-Zinn
(1990). As predicted, in the subsequent aggression task with the administration of
aversive noise blast, those in the mindfulness plus rejection condition showed less
aggressive behaviour than those in the rejection only condition, and their behaviour
16
did not differ from that of those in the non-rejection condition. The promising results
from Heppner et al.’s study indicated that provoked aggressive behaviour may be
reduced through a brief mindfulness induction. However, they did not include any
manipulation checks to confirm engagement in the mindfulness activity, and did not
provide information with regards to participants’ prior meditation experience.
Crucially, it is questionable whether the benefits of mindfulness can be separated
from people’s capacity of self-control (Masicampo & Baumeister, 2007). Review on
the role of self-control on aggression will be presented in the next section, followed
by our proposal for testing mindfulness in tandem with self-control on aggression.
1.2.3 Self-control and its impact on aggression
1.2.3.1 Definition and ingredients of self-control
Baumeister (1998, 2011) proposed that the “self” can be organised based on three
basic experiences. First, the self is a knowledge structure (a knower and a known),
which develops based on the experience of reflexive awareness of self (e.g., self-
esteem, self-perception, self-consciousness, self-monitoring, self-focused attention,
self-reflectiveness). Second, self is an interpersonal being (a belonger), which
involves interpersonal processes (e.g., self-presentation). Third, self is an agent with
an executive function (a doer), which makes choices and decisions, exerts control
over the environment, and regulates its own responses.
The capacity of the self to regulate itself by altering or overriding its dominant
response tendencies is known as self-control (Bandura, 1989; Baumeister,
Heatherton & Tice, 1994; Carver & Scheier, 1982, 2011). Because self-control
involves restraining impulses, researchers often treat the measures of self-control as
though they are equivalent with impulsivity; however, these are two independent
constructs (Duckworth & Kern, 2011).
The term self-control has also been used interchangeably with self-regulation, self-
discipline, willpower, effortful control, ego strength, and inhibitory control
17
(Duckworth & Kern, 2011). To avoid terminological confusion, it has been suggested
to allocate self-control as the deliberate, conscious, effortful subset of self-regulation
(Baumeister, Vohs, & Tice, 2007). As we have previously discussed (see section
1.2.2.2), mindfulness can also be seen as a conscious “subset” of self-regulation, but
a less effortful one. In this thesis, we use the terms self-controlled regulation and
mindful regulation whenever it is necessary to contrast these constructs.
According to the influential theory of feedback loops (Carver & Scheier, 1982, 2011),
originally developed by cybernetics theorists, self-control is a regulatory cycle
consisting of a sequence of four steps, under the acronym of TOTE (test–operate–
test–exit). In the testing phase, people are comparing the self or aspect of the self
against a reference value or standard of comparison (the so-called self-awareness).
Some operations to change the self can be initiated if a discrepancy between the
desired and current states is detected. Further tests are then performed to evaluate
whether the self has now been brought into line with the standard. When no further
operations are required to eliminate or reduce the discrepancy, the self-control
process enters the exit phase. During the testing phases, aversive emotions typically
arise when the self is below the relevant standard, whereas positive emotions occur
if the self has surpassed the standard (Baumeister, Schmeichel, & Vohs, 2007).
As the feedback loop theory implies, effective self-control consists of reducing the
discrepancy between the desired and current states. To do so, the desired states
must be set against a clear and well-defined standard. The standard can be some
ideals, values, morals, social expectations, or long-term personal goals. Apart from
standards, effective self-control requires monitoring (continuously observing one’s
current states or performance against the desirable goals or standards), the capacity
of the self to make changes, and motivation (Baumeister, Schmeichel, et al., 2007).
This thesis focuses on two ingredients of self-control: monitoring and the capacity of
the self. As we will argue in the latter section, the monitoring process, which
corresponds to the “Testing” phase in the TOTE theory, can be enhanced through
mindfulness. This may contribute to a more effective regulation of aggression.
18
1.2.3.2 Strength model of self-control
Given that research on self-control has focused on people’s effort to stimulate
desirable responses and inhibit undesirable responses, it is not surprising that body
of evidence has linked good self-control to a variety of benefits. Amongst these
benefits include healthier interpersonal relationship, greater popularity, better
mental health, effective coping skills, superior academic performance, as well as less
vulnerability to drug and alcohol abuse, aggression, criminality, and eating disorders
(see de Ridder, Lensvelt-Mulders, Finkenauer, Stok, & Baumeister, 2012; Moffitt et
al., 2011). However, exerting self-control is not without cost.
In considering the costs of self-control, it is relevant to differentiate between its trait
and state aspects. Trait self-control is a broad, dispositional tendency to exert self-
control; the state is a current propensity to exert self-control. While trait self-control
appears to be a highly desirable quality, the momentary act of self-control seems to
require significant sacrifices (for an overview, see Baumeister & Alquist, 2009).
According to Baumeister et al. (1994), much like a muscle, the capacity of the self to
change itself relies on the availability of a common and limited pool of self-control
“resources”, or self-control strength. The construct of self-control strength
corresponds to the “Operate” phase in the TOTE theory (Bauen & Baumeister, 2011;
Baumeister, Schmeichel, et al., 2007), when a person has detected a discrepancy
between the desired and current states, and is prepared to initiate actions to
eliminate or reduce the discrepancy. At this point, one must have sufficient inner
resources to alter its own responses. Because the common resource for self-control
is limited, after exerting initial self-controlled regulation, the capacity for further
control is decreased. This temporary deficit is known as ego-depletion.
Early evidence for ego-depletion was reported from Baumeister and colleagues
laboratory (i.e., Muraven, Tice, & Baumeister, 1998; Baumeister, Bratslavsky,
Muraven, & Tice, 1998). In one study, individuals who were required to regulate
emotions while watching an upsetting video performed less well on subsequent
handgrip stamina. In their other study, when participants were required to suppress
thoughts about a white bear, they subsequently more quickly gave up on an
19
unsolvable anagram. In these studies, controlling aspects of the self that are
normally spontaneous and impulsive, such as expressing emotions or the innate
tendency to recall thoughts of a suggested object (the so-called cognitive rebound
phenomenon) requires effortful self-control and draws from a reservoir of self-
control resources. In doing so, it is argued that the resource is depleted, leaving less
available resources to exert self-control on subsequent self-control tasks.
Following Baumeister and colleagues’ initial tests, studies have replicated the ego-
depletion effect extensively. Various spheres of self-control have been employed in
laboratory settings, including controlling attention, controlling emotions, controlling
impulses, controlling thoughts, cognitive processing, choice and volition, and social
processing (Baumeister, Vohs, et al., 2007). Supporting the strength model proposal
on a common resource of self-control, Hagger, Wood, Stiff, and Chatzisarantis’s
(2010) meta-analysis found a robust effect of depletion on performance on self-
control task, with minimal variation across spheres of self-control.
While performing self-control acts may temporarily deplete the self-control
resources, repeated practice of self-control over a period of time may enlarge or
strengthen these resources. This effect has been shown particularly in the form of a
less susceptibility to depletion. For instance, in Muraven, Baumeister, and Tice’s
(1999) study, participants’ self-control performance was measured using a task of
overriding physical discomfort (a handgrip task), given before and after an ego-
depleting task (suppressing thoughts of a white bear) prior to a self-control training
of tracking eating habits or regulating posture for two weeks, and following self-
control training. As expected, following the self-control training, participants
displayed better handgrip performance after the thoughts suppression task, as
opposed to their pre-training handgrip performance.
Likewise, people’s ability to regulate their use of stereotypes improved following two
weeks of self-control training of using the nondominant hand or avoiding
colloquialisms/cursing (Gailliot, Plant, Butz, & Baumeister, 2007). In a series of
studies, Oaten and Cheng also showed that people’s ability on the visual tracking
task increased after they were trained over a period of months in formal academic
20
study (2006a), physical exercise (2006b), and financial monitoring (2007). Crucially,
Muraven (2010) demonstrated that effect of self-control training on self-control
performance following depletion cannot be explained by alternative mechanisms
such as self-awareness (i.e., monitoring one’s self-control activities with no specific
attempt to practice self-control), self-fulfilling prophecies, or self-efficacy.
In short, the existing research appears to establish a direct relationship between self-
control and ego-depletion such that greater exertion of self-control leads to more
depletion. Until the resource for self-control can be replenished, one’s ability to
perform further self-control acts is compromised. Several counteracting variables for
the depletion effect on self-control performance have been identified, such as
glucose, humour, laughter, cash incentives, implementation intentions, and social
goals (see Baumeister, Vohs, et al., 2007), and more recently, mindfulness (Friese,
Messner, & Schaffner, 2012). Just as a muscle that gets stronger through exercise, a
longer term effect of self-control practice also promotes a greater resistance to
depletion (see Baumeister, Gailliot, DeWall, & Oaten, 2006).
1.2.3.3 Self-control and aggression
In Gottfredson and Hirschi’s (1990) general theory of crime, deficient in self-control
has been acknowledged as a leading factor in understanding criminality and violent
activity. In fact, individual differences in self-control have been shown to predict
rates of behavioural problems and criminality over life-course development (Caspi,
2000; Moffitt et al., 2011). Sex differences in aggression have also been linked to
differences in self-control between males and females (Burton, Cullen, Evans, Alarid,
& Dunaway, 1998). Cognitive neuroscience models of self-control also demonstrate
that self-control failure occurs when strong impulses overwhelm prefrontal cortical
control mechanisms, and deficits in the function of prefrontal regions has been
shown to predict aggression and violence (e.g., Heatherton & Wagner, 2011).
Self-control also seems to be related to the strategies people choose for managing
and expressing their hostile thoughts and anger. Individuals high in self-control
21
reported lower outwardly directed aggression and less inclination to ruminate about
their anger, as opposed to those low in self-control (Tangney, Baumeister, & Boone,
2004). Moreover, hostile thoughts and anger are more likely to manifest as
aggressive behaviour when people’s cognitive inhibition has been impaired (see
Giancola, 2000). These findings suggest that individuals who lack the ability to
control their anger are more likely to engage in aggressive behaviour.
DeWall, Baumeister, Stillman, and Gailliot (2007) provided a series of experiments on
how ego-depletion may affect aggression (see also Denson, von Hippel, Kemp, &
Teo, 2010; Finkel, DeWall, Slotter, Oaten, & Foshee, 2009; Stucke & Baumeister,
2006; Vohs, Glass, Maddox, & Markman, 2011). In one of DeWall et al.’s study,
participants were allocated to a depletion condition (resisting the urge to eat a
tempting donut) or non-depletion condition (resisting the urge to eat a radish). All
participants then received an insulting feedback of their essay from an ostensible
partner. Participants were then given an opportunity to prepare a spicy snack for
their partner (whom had indicated a strong dislike for spicy foods). As expected,
depleted participants gave more hot sauce to the partner than non-depleted
participants. In subsequent studies, depleted participants showed higher aggressive
responses (i.e., aversive noise blasts or negative job candidate evaluation) only when
they had been insulted, but not in the absence of insulting provocation. In addition,
the intention to aggress was less expressed by those high in trait self-control.
Since repeated practice of self-control has been shown to increase resistance to
depletion, it may also decrease aggression. Empirical support for this prediction was
reported by Denson, Capper, Oaten, Friese, and Schofield (2011; see also Finkel et
al., 2009). In the first session of Denson et al.’s study, self-control trained
participants were asked to use their nondominant hand in mundane tasks (e.g.,
brushing their teeth, opening doors, operating a computer mouse) for a 2-week
period, exerting as much control as possible when performing these tasks. In the
following laboratory session, both self-control and non-self-control trained
participants received an insulting feedback from an ostensible partner, before they
competed in an aversive noise blasts game with this partner. The results indicated
that practicing self-control reduced aggressive behaviour in particular for those high
22
in trait physical aggression. Self-control trained participants also reported less angry
rumination affect, even after controlling for trait anger, suggesting that self-control
training may reduce aggression by facilitating self-control capacity to regulate anger.
The aforementioned research suggests that while depletion may not directly
increase aggression, it limits self-control ability from acting on aggressive impulses
when the urge to aggress is relatively strong (e.g., after receiving insulting feedback).
This supports the strength model of self-control argument’s that the proximal
antecedent of aggression is often a failure of self-control (for an overview, see
Denson, DeWall, & Finkel, 2012; DeWall, Finkel, & Denson, 2011). To our knowledge,
whether or the effect of depletion on aggression can be counteracted by other
variables has not been reported. In this thesis, we examine whether mindfulness
may have the potential to attenuate such effect.
In the same way, studies examining the effect of self-control training on aggression
typically involve the presence of provocation, either given following depletion (Finkel
et al., 2009) or in the absence of depletion (Denson et al., 2011). It appears that the
long-term effect of practicing self-control may not also directly reduce aggression
unless people encounter some situations that trigger aggressive responding. To
suggest evidence on the specific contribution of mindfulness on aggression
independent of self-control mechanisms, mindfulness should also uniquely reduce
provoked and unprovoked aggression in the absence of depletion condition.
1.2.4 Relating mindfulness, self-control, and aggression
1.2.4.1 Distinguishing between mindfulness and self-control
Before building on a proposal for the plausible ways for testing the role of
mindfulness and self-control conjointly on aggression, it is important to distinguish
between mindfulness and self-control on conceptual levels.
Brown et al. (2007a, 2007b) have provided a thorough inspection on the differences
between mindfulness and self-control, using the notion of “self” as a vantage point.
23
Fundamentally, there are two basic functions of consciousness (see also Western,
2001): monitoring experiences as they unfold (an “observer” function) and
controlling the content of consciousness (a goal-directed agent of maintenance and
change function). From the perspective of the feedback loop theory (Carver &
Scheier, 1982; 2011), as well as other self-awareness theories, these functions are
intertwined. The self as an agent with an executive function (Baumeister, 1998;
2011) exerts control over itself in order to achieve some future goals. For this
purpose, self-controlled attention demands the monitoring of internal and external
realities prior to and during control efforts. Attention is, therefore, goal-oriented,
and mostly directed to aspects of the self through the experience of reflexive
awareness (self-reflection).
Mindfulness serves a monitoring function of consciousness (Brown et al., 2007a,
2007b). This implies that with mindfulness, the two basic functions of consciousness
can be disentangled. Since mindful attention is given on the task at hand, it allows
one to fully experience one’s own content of consciousness (i.e., thoughts, emotions,
body sensations, values, plans, memory), in a situation of exposure (e.g., Baer, 2003;
Linehan et al., 2007; Shapiro et al., 2006). Through exposure, one learns that the
content of consciousness can be observed as a transitory state that is arising and
passing. Sustained, nonjudgmental observation of any uncomfortable content of
consciousness (e.g., anger, anxiety, pain), with minimal effort to suppress, avoid, or
act on them should eventually lead to “the extinction of fear responses and
avoidance behaviours previously elicited by these stimuli” (Baer, 2003, p.5).
As Brown and colleagues noted, the result of this process is a mental gap between
attention and its objects, or between consciousness and its contents. A similar
process has been acknowledged by Shapiro et al. (2006) as a fundamental shift in
perspective or reperceiving. In this manner, mindfulness enables one to change the
relationship to the consciousness and its content (e.g., S. C. Hayes et al., 2011; Segal
et al., 2012), rather than altering the content of consciousness through self-
controlled efforts.
24
Mindfully monitoring the “in-the-moment” experience also includes being aware of
any thinking that one does about the future goals (and the past). Thus even though
mindfulness is conceptually different from self-control, it is not necessarily
antithetical to the goals of the self (Brown et al., 2007a). Research has documented a
positive relation between individuals’ tendency to be mindful and their reported
ability to self-regulate (e.g., Barnes et al., 2007; Brown & Ryan, 2003; Fetterman,
Robinson, Ode, & Gordon, 2010; Lakey, Campbell, Brown, & Goodie, 2007). Crucially,
both mindfulness and self-control may facilitate adaptive responding to the
environment, as opposed to the more habitual behaviour patterns. As presented in
the two previous sections, the benefits of these two processes are found to be
similar, including reductions in aggression.
Nevertheless, it is likely that the subjective outcome of mindfulness and self-control
would different. Self-controlled regulation is depleting (see section 1.2.3.2),
particularly in the short term, whereas mindful regulation appears to be vitalising
and energising (Brown & Ryan, 2003; Williams & Penman, 2011). A series of studies
(Brown, Kasser, Linley, & Ryan, 2009) showed that regardless of economic status,
mindful individuals and those who participated in mindfulness meditation training
reported a smaller financial desire discrepancy, which explained their higher levels of
subjective well-being (affective state and life satisfaction). Thus mindfulness may
promote the perception of being content that what one has is “enough”.
There is also empirical support that trait mindfulness and participation in a
mindfulness-based intervention (i) reduces the self-reported frequency of negative
automatic thoughts about the self, and (ii) increases the ability to let go of these
thoughts (Frewen, Evans, Maraj, Dozois, & Partridge, 2008). The combinations
between the capacity to let go of negative thoughts and the perception of having
enough (as opposed to wanting more) may therefore enable individuals to increase
the degree of observation on the task at hand. The mindful capacity of observation
also includes observing one’s own attempt to exert self-control; thus may facilitate
better choices about whether and when to control the self in ways that are more
congruent with one’s actual values (Brown et al., 2007a; Shapiro et al., 2006).
Indeed, it has been suggested elsewhere that adaptive self-regulation should also
25
incorporate the capacity to disengage from unattainable goals as well as to re-
engage effort and commitment towards pursuing more meaningful goals (Wrosch,
Scheier, Carver, & Schulz, 2003; Wrosch, Scheier, Miller, Schulz, & Carver, 2003).
With these distinctions in mind, we now turn into a discussion on our proposal for
investigating the role of mindfulness and self-control on aggression.
1.2.4.2 Framework for mindfulness, self-control, and aggression
Empirical attempts to clarify the mechanism by which mindfulness may reduce
aggression were conducted in Borders et al.’s (2010) correlational study in both
undergraduates and non-undergraduates. Specifically, they proposed rumination
(i.e., a process characterised by internal, sustained, and rigid attention to perceived
discrepancies between current and desired states) as a mediator on the mindfulness
and aggression link. Mindfulness has been suggested to break the repetive cycle of
rumination by re-directing attention to the present (Shapiro, Oman, Thoresen,
Plante, & Flinders, 2008). Borders et al. found that rumination partially mediated the
link between mindfulness and anger and hostility but only amongst undergraduates.
Thus, while rumination may be a chief mechanism between trait mindfulness and
the emotional and cognitive components of aggression, other mechanisms may
come into play for reductions of the behavioural components of aggression.
Accordingly, Borders et al. proposed that the effect that mindfulness has on
aggressive behaviour may been mediated through relaxation, emotion regulation
(e.g., altering thoughts or behaviour to cope with negative emotions), better
cognitive functioning and flexibility (i.e., self-regulation of attention and cognitive
inhibition), and decrease impulsivity. Some of these suggested mechanisms are
related to self-control. Self-control was explicitly mentioned (but not tested) by
Heppner et al. (2008) as a potential mediator between mindfulness and aggression.
As previously discussed, mindfulness may facilitate the self’s ability to exert control
in accordance to one’s actual values (Brown et al., 2007a; Shapiro et al., 2006). Thus
it is plausible that mindfulness may increase self-control capacity, and in turn
26
increases in self-control capacity would reduce aggression. At the same time, to
clarify its distinct contribution, mindfulness should also produce additional reduction
in aggression after accounting for self-control. There are several ways to examine
our proposal that mindfulness may influence aggression in the same way it is
proposed to be influenced by self-control, and that the effect of mindfulness may
even be independent of and beyond that of self-control.
Firstly, the potential role of mindfulness and self-control on aggression may be
investigated in the trait level. When encountering a situation that potentially could
trigger aggressive responses (e.g., provocation), mindful individuals should be aware
of their emotions (e.g., anger), but also should be less judgemental and reactive.
Rather then being identified as an integral part of the self that needs to be acted
upon, these emotions are being observed or witnessed as a state that is arising and
passing, but not necessarily needing action. As described earlier (see section 1.2.4.1),
keeping attention on emotions may lead to exposure and, in turn, extinction.
Consequently, mindful individuals may experience the resulting aggression-related
emotions to a reduced extent.
Individuals who are higher in self-control (e.g., discipline, hard-working, reliable)
should be able to refrain from acting on aggressive impulses through their capacity
of altering, overriding, or manipulating aggression-related emotions. However, they
may be lacking in the awareness and acceptance of these emotions. When emotion
acceptance is difficult, these individuals may try to avoid or suppress their emotions
altogether. The paradoxical effect of emotion avoidance and suppression strategies
in increasing the intensity and frequency of distressing thoughts and feelings has
been well-documented (see e.g., Gross, 2002; Gratz & Roemer, 2004).
Moreover, difficulty in accepting emotions may activate the link between these
emotions and aggression. For example, male’s aggressive behaviour has been shown
to be related to restrictive emotionality (presumably due to traditional sex roles),
particularly emotional non-acceptance, rather than to the overall inability to self-
regulate (Cohn, Jakupcak, Seibert, Hildebrandt, & Zeichner, 2010). In a current
overview, Roberton, Daffern, and Bucks (2012) suggest that over-regulating
27
emotions may lead to aggressive behaviour by increasing negative affect, reducing
inhibitions against aggression, compromising decision making processes, diminishing
social networks, increasing physiological arousal, and hindering the resolution of
difficult situations. Thus, even though self-control ability can decrease aggression
and anger, mindfulness may arguably provide a more adaptive alternative. Indeed, a
recent correlational study showed that trait mindfulness predicted incremental
variance over trait self-control in psychological well-being and general distress
(Bowlin & Baer, 2012).
Secondly, it is plausible to test the effect of state induced mindfulness on aggressive
behaviour under ego-depleted condition. Self-control is often initiated when
discrepancies between desired states and current states are detected (Carver &
Scheier, 1982, 2011). Self-monitoring is also central to Novaco’s (2003) conception of
anger regulation. Inzlich and Gutsell (2007) demonstrated that after performing an
initial self-control act, people’s neural system that monitors the discrepancy
between desired and current state (error-related negativity [ERN]) is weakened,
suggesting that depletion limits the attentional system for initiating self-control acts.
Other potential mechanisms have also been identified as mediating variables for
depletion effect on self-control performance, such as effort, perceived difficulty,
negative affect, subjective fatigue, and blood glucose level—indicating the effortful,
aversive nature of self-control tasks (Hagger et al., 2010). Alternatively, depleted
people may become less motivated to exert self-control because they want to
conserve their energy for more important future tasks (Tyler & Burns, 2009). These
depleted people may then behave aggressively in particular when they encounter
subsequent aggression-provoking events (see Denson et al., 2012; DeWall et al.,
2011). By mindfully exposing oneself to the uncomfortable emotions, thoughts, or
body sensations associated with depletion and provocation, one should be more
tolerant to and less bothered by such experiences. This flourishing effect of
mindfulness has been shown to counteract the effect of depletion on self-control
performance by Friese et al. (2012).
28
Thirdly, laboratory study of mindfulness, self-control, and aggression may also be
conducted in the absence of depletion. Extensive training in mindfulness (i.e.,
mindfulness-based interventions) has been shown to reduce aggression (see Fix &
Fix, 2013; Shonin et al., in press). However, the similarities between features of
mindfulness-based interventions and training in self-control have been highlighted in
Masicampo and Baumeister’s (2007) commentary. Specifically, both training involve
regulation of one’s thoughts and behaviours, adherence to exercises, and
commitment to the exercises over extended periods. Therefore, mindfulness might
well be argued to be simply one example of self-control training.
As a comparison, self-control training typically lasts for two weeks, compromising a
regular practice of small acts of altering one’s dominant response tendencies in daily
activities such trying to improve one’s posture, tracking eating habit, using
nondominant hand in a mundane task, and avoiding colloquialisms/cursing (see
Baumeister et al., 2006). By contrast, the accepted standard format for the MBSR
programme is eight weeks of a total 26 hours class time (Carmody & Baer, 2009),
where participants practice various meditation skills (i.e., body scan, hatha yoga,
mindfulness of breathing, mindfulness of body sensations), along with discussion of
stress, coping, and homework assignments, in addition to practising mindfulness
during ordinary activities such as walking, standing, and eating (see Baer, 2003).
Given the complexity of the clinical package of mindfulness, the first few weeks has
been reported to be especially difficult for novice practitioners (Segal et al., 2012). As
speculated elsewhere, during the initial stages of mindfulness practice, cognitive
control may be needed over the emotional reactions to assist attention on the task
at hand (Holzel et al., 2011; Sauer et al., 2013); thus imposing a high degree of self-
control exercises. However, as Brown et al. (2007a) argued, the aim of mindfulness
practice is not to achieve the goals and demands of the self, but to provide a “space
for awareness” to open. Unfolding the point at which this space of awareness could
afford an independent effect of mindfulness over self-control is of particular
interesting. To do so, the state of mindfulness should be able to strengthen the
effect of self-control when the self-control resource has not been depleted, as well
as when this resource has been bolstered (via self-control training).
29
1.2.4.3 Role of individual differences in sensitivity to provocations and
frustrations, and self-harm
As proposed by Lawrence (2006), assessment on individual differences in SP and SF
may generate specific predictions about who is likely to respond aggressively to
which triggers. Lawrence and Hutchinson’s (2013b) recent study showed that after
accounting for trait aggression, individuals high in SP behaved more aggressively
when provoked but not when unprovoked. However, the moderating role of SP on
the link between provocation and aggressive behaviour disappeared when
participants were asked to deal with an initially non-aggressive partner. As such,
Lawrence and Hutchinson suggested that it is plausible for those who are more
sensitive to provocation to be more responsive not only to signs of provocation, but
also to the absence of signs of aggression when aggression is expected (i.e., a
partner who has the opportunity to behave aggressively but initially refrained from
doing so).
As both mindfulness and self-controlled regulations may influence the link between
aggressive impulses and aggressive behaviour, individual high in SP may also be of
particularly responsive to these regulation strategies. Given that this thesis examines
aggressive behaviour in response to provoked and unprovoked aggression, a similar
effect may also occur amongst those high in trait aggression, although would
probably less salient in those high in SF. Measuring individual differences in
aggression, SP, and SF would allow us to examine to what extent mindfulness and
self-control may attenuate aggression amongst individuals who are already
predisposed to behave aggressively.
In this thesis, the role of mindfulness and self-control on aggression is also explored
while accounting for individual differences in self-harm. While the literatures on
aggression and self-harm are commonly separated, the presence of similar possible
antecedents for self-harm and aggression (see section 1.2.1.2) indicates that these
two harmful behaviour might be related.
Current self-harm theories appear to share the view that individuals who engage in
self-harm do so partly to avoid, escape, manage, or regulate their very intense
30
emotions (e.g., Chapman, Gratz, & Brown, 2006; Gratz & Roemer, 2004; Linehan,
1993; Nock & Prinstein, 2004). Although there are other forms of aversive
experiences (e.g., thoughts, memories, or body sensations), these too are likely to be
unwanted due to the emotions accompanying them (Chapman et al., 2006).
Similarly, aggressive behaviour has also been conceptualised as maladaptive
emotion regulation strategy (see Roberton et al., 2012).
With the inclusion of mindfulness into the third-wave cognitive-behavioural
approaches of self-harm, individuals are thought to become less avoidance to their
very intense emotions, which may prevent repeated episodes of self-harm (Linehan,
1993; S. C. Hayes et al., 2011; Williams, Dugan, Crane, & Fennell, 2006). However,
studies on emotion regulation difficulties amongst self-harmers have yet clarified the
extent to which mindful and self-controlled emotion regulation may independently
contribute to reductions in self-harm (see Slee, 2008).
1.3 RESEARCH QUESTIONS AND THESIS OVERVIEW
To summarise, in the light of the literature reviewed so far, the primary question
addressed by the thesis is: “What is the effect of mindfulness on aggression?” In
addition, the potential role of self-control on the potential link between mindfulness
and aggression is investigated.
These research questions are elaborated through a series of studies. Study 1,
presented in the following chapter, is a psychometric study exploring the
relationships between mindfulness, self-control, and aggression (both to self and
others) in its dispositional forms. Study 2 is an experimental study comparing the
impact of mindfulness induction and self-control training on provoked and
unprovoked aggressive behaviour. Study 3 is an experimental study exploring the
role of mindfulness induction on provoked and unprovoked aggressive behaviour,
under the condition of self-control ego-depletion. Given that the concept of
mindfulness originates in Eastern contemplative traditions, it is necessary to test the
31
effect of mindfulness beyond the Western sample. Therefore, study 4 is a replication
of Study 3, using a cross-cultural sample from Indonesia.
In this way, we will be able to suggest evidence for the unique contribution of
mindfulness on aggression apart from the mechanism of self-control, and seek
whether this potential effect varies across cultures. In addition, all studies in
reported in this thesis also test the role of individual differences in aggression
(physical aggression, verbal aggression, anger, hostility), sensitivity to provocations
(SP) and frustrations (SF), and self-harm on the potential link between mindfulness,
self-control, and aggression.
32
CHAPTER TWO
Study 1: Relationships between trait mindfulness, self-control, and aggression
2.1 INTRODUCTION
The current study explores the relationships between mindfulness, self-control, and
aggression in their dispositional form. In addition, we assess the extent to which trait
mindfulness and self-control may influence (i) self-harm in the same way as they may
influence aggression, and (ii) the potential link between individual differences in
sensitivity to provocation (SP) and frustration (SF) and aggression/self-harm.
2.1.1 Individual differences in mindfulness, self-control, and aggression
An increasing amount of published work has pointed out that mindfulness may
attenuate the associations between aggressive triggers and the difficult emotions
experienced (see e.g., Wright et al., 2009 for a review). That is, although all
individuals may encounter events in the environment that potentially could trigger
aggressive responses, those who are mindful may experience the resulting negative
emotions (e.g., anger) to a reduced extent. When conceptualised as a dispositional
variable, mindfulness is associated with lower levels of self-reported physical and
verbal aggression (Borders et al., 2010; Heppner et al., 2008, Kelly & Lambert, 2012),
and with anger and hostility (Barnes et al., 2007; Borders et al., 2010; Brown & Ryan,
2003; Kelly & Lambert, 2012).
Some authors have theoretically suggested self-control (Border et al., 2010; Heppner
et al., 2008) as one of the potential mediators between mindfulness and aggression.
The link between self-control and aggression per se has been well documented (e.g.,
Caspi, 200; Gottfredson & Hirschi, 1990; Moffitt et al., 2011), and self-reported
measures of mindfulness and self-control are strongly related with each other
(Barnes et al., 2007; Brown & Ryan, 2003; Fetterman et al., 2010; Lakey et al., 2007).
However, the differential association between mindfulness, self-control, and
33
aggression has not been reported. Therefore the primary aim of the current study is
to explore the relationships between these variables in their dispositional forms.
Outside the area of aggression, Parto and Besharat’s (2011) correlational study using
high risk males from public high schools in Tehran has demonstrated the mediating
role of self-control (as measured by the Self-Regulation Inventory [SRI -25]: Ibanez,
Ruipérez, Moya, Marqués, & Ortet, 2008) on the link between mindfulness
(Philadelphia Mindfulness Scale [PHLMS]: Cardaciotto et al., 2008) and psychological
well-being. Specifically, they found that despite the direct effect of mindfulness on
the psychological constructs of well-being and distress, mindfulness influenced self-
control and in turn self-control affected psychological well-being. In the current
study, we test whether a similar mediation model of trait self-control also applies on
the association between trait mindfulness and trait aggression.
As discussed in Chapter 1, it is still questionable whether the benefit of mindfulness
in predicting adaptive functioning can be separated from the more general self-
control mechanism (Masicampo & Baumeister, 2007). Recently, Bowlin and Baer
(2012) demonstrated that mindfulness (as measured by the Five Facet Mindfulness
Questionnaire [FFMQ]: Baer et al., 2006) indeed accounted for significant variance in
psychological well-being and general distress after accounting for self-control (Self
Control Scale [SCS]: Tangney et al. 2004). In the current study, we examine the
unique contribution of mindfulness on trait aggression after we control for the role
of trait self-control and individual differences in aggression.
Furthermore, Bowlin and Baer has also found a moderating role of mindfulness on
the link between self-control and general distress. Provided that mindfulness may
predict incremental variance in aggression, it may also be possible for mindfulness to
act as a moderator on the relationship between self-control and aggression. Self-
controlled individuals might inhibit their impulsive responses of behaving
aggressively, but lacking in the awareness and acceptance of aggression-related
emotions. This excessive control of emotions can, at some point, increase aggressive
behaviours (current review on the impact of deliberate emotion regulation on
aggression is provided in Roberton et al., 2012). Conversely, the presence of a
34
mindful quality would enable self-controlled individuals to choose to self-regulate in
healthier senses (Shapiro & Carlson, 2009), including whether and when they want to
exert control (Brown et al., 2007a), and therefore act less aggressively. The
moderating effect of trait mindfulness on the association between trait self-control
and trait aggression is also explored the current study.
2.1.2 Link between mindfulness, self control, aggression, and self-harm
A previous review (Hilbrand, 2001) has pointed out that aggression to others and
harm to the self (self-harm) often coexist; however, the risk assessment of these
behaviours are commonly separated. Conceptually, it could be argued that those
who self-harm are less likely to harm others, as some research has suggested that
self-harm is psychological distress, particularly anger, directed inwards (Hills &
Dallos, 2012). Repetition of self-harm was also associated with levels of intropunitive
hostility (hostility towards the self) and hopelessness, but not with extrapunitive
hostility and dominance (Brittlebank et al., 1990). Even so, the presence of similar
possible mechanisms for self and others harm (e.g., Hillbrand, 2001; Placidi et al.,
2001; Plutchick & van Praag, 1989; Roaldset et al., 2011), would mean that
aggressive individuals may, in fact, also lack of inhibition to harm themselves.
When dealing with high levels of negative emotions, individuals may ruminate about
the causes, situational factors, and consequences of these emotions, or use thought
suppression as an attempt to stop ruminating on these emotions (Selby, Anestis, &
Joiner, 2008). These self-controlled regulation strategies may deplete the resource of
self-control and thereby increasing the association between rumination and the
negative emotions (Denson, 2013). To break the vicious cycle between negative
emotions, rumination, and thought suppression, individuals may then engage in a
dysregulated behaviour (e.g., self-harm, aggression) in order to distract themselves
from these emotions (Selby, Franklin, Carson-Wong, & Rizvi, in press; see also
Roberton et al., 2012).
35
Indeed, emotional relief has been reported as the most common reason for self-
harm in studies using self-report methodologies (Brown, Comtois, & Linehan, 2002;
Chapman & Dixon-Gordon, 2007). Likewise, aggression and violence toward others
may serve as an affect regulatory function (Berkowitz, 1990; 2008; Bushman,
Baumeister, & Phillips, 2001; Jakupcak, Lisak, & Roemer, 2002). Ruminating about a
provocation was also shown to increase the likelihood of displaced aggression, even
following minor triggering events (Bushman, 2002; Bushman, Bonacci, Pederson,
Vasquez, & Miller, 2005).
By contrast, mindfulness may decrease both over-engagement (e.g., rumination) and
under-engagement (e.g., avoidance) of experiences, by bringing attention back with
a nonjudgemental attitude (Bishop et al., 2004; A. M. Hayes & Feldman, 2004;
Shapiro et al., 2006). The mindful lack of trying to automatically ruminate on or resist
negative emotions may result in less need to harm the self and others.
Gratz and colleagues (Gratz, 2007; Gratz & Roemer, 2004) have argued that clinical
definitions of emotion regulation should capture the awareness of and acceptance of
negative emotions, as well as the ability to engage in goal-directed behaviours and
inhibit impulsive behaviour. As such, the first two components are the core of
mindfulness (Bishop et al., 2004; Brown & Ryan, 2003; Kabat-Zinn, 1994); whereas
the latter resemble the spheres of self-control in self-control theory (Baumeister,
Vohs, et al., 2007; Hagger et al., 2010; Tangney et al., 2004).
Few published studies that have investigated the association between specific
components of emotion regulation and aggression/self-harm seem to yield
inconsistent results. Gratz and Roemer’s (2004) study reported no significant
differences between components of emotion regulation and self-harming behaviour
a non-clinical population. However, the frequency of partner abuse in their study
was predicted by the self-control related components only. Another study (Slee,
Garnefski, Spinhoven, & Arensman, 2008) showed that mindfulness-related and self-
control related components independently distinguished between clinical self-
harmers and non self-harmers. Yet another study (Slee, Spinhoven, Garnefski, &
Arensman, 2008) found that only difficulties in the self-control related component
36
mediated the treatment effect on self-harming behaviour amongst clinical
population. In the current study, we examine the role of mindfulness on self-harm
after we controlled for the influence of self-control.
Furthermore, levels of anxiety and depression are frequently implicated in the
literature examining the etiology of self-harm (see Hamza, Stewart, & Willoughby,
2012). Hence we also examine the impact of mindfulness and self-control on self-
harm while controlling for the effect of anxiety and depression on self-harm.
It is also plausible, therefore, that mindfulness may moderate any relationships
between self-control and self-harm in the same way it might moderate the link
between self-control and aggression. Specifically, individuals who higher in both
mindfulness and self-control might be less likely to self-harm compare to those who
are higher in self-control but lower in mindfulness.
2.1.3 Individual differences in sensitivity to provocations and frustrations
General trait aggression is not the only factor determining why individuals may
become aggressive in any particular situation. While person-based and situation-
based antecedents of aggression are theoretically linked (see Anderson & Bushman’s
General Aggression Model, 2002), however, these factors are typically investigated
separately (Lawrence, 2006).
As discussed in the previous chapter, Lawrence and colleagues (Lawrence, 2006;
Lawrence & Hodgkins, 2009; Lawrence & Hutchinson 2013b) recommend examining
individual differences in sensitivity to provocations (SP) and frustrations (SF) to
generate a clear set of specific predictions about who is likely to respond
aggressively to which triggers. Distinct relationships was shown between these traits
and subscales of trait aggression (as measured with Buss & Perry’s Aggression
Questionnaire, 1992), in which SP was related to behavioural or overt aggression
whereas SF to hostility and anger (Lawrence, 2006). This highlights the importance
for the current study to scrutinise the potential effect of SP and SF on the link
between mindfulness, self-control, and trait aggression.
37
Whilst provocations and frustrations are assumed to operate similarly by activating a
network linking aggression-related cognitions, emotions, and behaviours (according
to cognitive neoassociation theory, e.g., Berkowitz, 1990), mindfulness and self-
control may arguably reduce the activation of these links, thereby decreasing the
likelihood that individuals will act aggressively. Specifically, individuals who are high
in SP and SF but at the same time either mindful (allow the negative emotions to be
experienced and not acted upon), or self-controlled (those who experienced the
aggressive triggers but do engage in self-control to resist behaving aggressively)
should score lower in aggression. The proposed moderating role of mindfulness and
self-control on the association between SP and SF and aggression would also be
explored in this study.
A similar moderation model is also plausible for the prediction of self-harm. Apart
from anxiety and depression, emotional vulnerability in the form of emotional
reactivity (i.e., high sensitivity to emotional stimuli) and emotional intensity (i.e., the
tendency to have extreme reactions) has been suggested as one of the individual risk
factors for self-harm (Linehan, 1993). It is therefore conceivable that individuals who
are more sensitive to provocations and frustrations triggers will be more likely to
have engaged in self-harmful behaviours. In this sense, mindfulness and self-control
might moderate any relationships between SP and SF and self-harm in the same way
they moderate the link between these variables and aggression.
In sum, recent literature suggests two different perspectives on the relationships
between mindfulness, self-control, and aggression. The first possibility is that self-
control may mediate the relationship between mindfulness and aggression. The
second one is that mindfulness may moderate the relationship between self-control
and aggression. Testing both possibilities is of particular interest of the current
study. Additionally, we explore whether (i) similar mediation and moderation models
also predict individual differences in self-harm, and (ii) mindfulness and self-control
moderate the effect of sensitivity to provocations and frustrations on
aggression/self-harm.
38
2.2 HYPOTHESES
The current study examines the following primary hypotheses:
1. Trait mindfulness and trait self-control will be positively correlated to each other,
negatively associated with trait aggression.
2. Trait self-control will mediate any relationships between trait mindfulness and
aggression.
3. Trait mindfulness will account for significant variance in trait aggression after
accounting for the potential influence of trait self-control.
4. Trait mindfulness will moderate the link between trait self-control and trait
aggression.
The first set of secondary hypotheses concern the possible link between trait
aggression and self-harm, and predicts that:
5. Individual differences in aggression will be positively associated with self-harmful
behaviours.
6. Those who score higher in mindfulness and self-control will be less likely to self-
harm, thus mirroring the associations with trait aggression.
7. Trait self-control will mediate any relationships between trait mindfulness and
self-harm.
8. Trait mindfulness will account for significant variance in self-harm after
controlling for:
a. Trait self-control.
b. Anxiety and depression.
9. Trait mindfulness will moderate the link between trait self-control and self-harm.
The next set of secondary hypotheses concern the role of individual differences in
sensitivity to provocations (SP) and frustrations (SF), and predicts that:
10. Individual differences in SP and SF will be positively associated with trait
aggression and self-harm.
39
11. Those who score higher in SP and SF, while at the same time are also mindful or
self-controlled, will be less likely to:
a. Harm others.
b. Harm themselves.
We predict that individuals who are more mindful and more self-controlled will be
less aggressive and self-harmless. Self-control will mediate the relationship between
mindfulness and aggression/self-harm. Moreover, mindfulness will account for
significant variance in aggression/self-harm after accounting for self-control and
other covariates. Mindfulness will moderate the relationship between self-control
and aggression/self-harm. Finally, both mindfulness and self-control will moderate
any relationships between SP and SF and aggression/self-harm.
2.3 METHODS
2.3.1 Participants
Participants were recruited from the end of March until the middle of April 2011. To
capture a wide range of data, we recruited general population (i.e., both university
and non-university students) with no exclusion criteria. Due to extensive missing
data, we removed 68 subjects out of the total 309 participants who agreed to
participate in the survey (see section 2.3.7 for explanation of participants’ removal).
Our final sample consisted of 241 subjects (152 females, 4 did not report sex). Ages
ranged from 18 to 41 (M = 23.87, SD = 6.00), 87.6% participants were university
students. Just over 67.6% participants were White, 24.9% were Asian, 4.6%
participants rated themselves as belonging to “Others” ethnical background, and
3.4% did not provide information. Concerning current mindfulness practice, 88.8%
participants reported never having encountered mindfulness, 5.4% practised them
once a year, 2.5% on monthly basis, 3.3% on weekly basis, and none practised them
on daily basis.
40
2.3.2 Design
An online survey using a surveymonkey website was conducted in which participants
completed the same set of questionnaires (Appendix 2.1) of individual differences in
mindfulness (Mindfulness Attention Awareness Scale [MAAS]: Brown & Ryan, 2003),
self-control (Self-Control Scale [SCS]: Tangney et al., 2004), aggression (Aggression
Questionnaire [AQ]: Buss & Perry, 1992), sensitivity to frustrations and provocations
(Situational Triggers of Aggressive Responses [STAR] scale: Lawrence, 2006), self-
harm (Deliberate Self-Harm Inventory [DSHI]: Gratz, 2001), and depression and
anxiety (Hospital Anxiety and Depression [HADS]: Zigmond & Snaith, 1983).
The online version of questionnaires was chosen not only to allow a relatively easy
access of potential participants at low cost, but also to minimise participant’s
barriers in reporting relatively sensitive or socially undesirable behaviours such as
aggression and self-harm. As shown by Metzger et al. (2000), similar types of
computer-based interviews reduced under-reporting of sensitive information, which
might otherwise occur due to participant’s embarrassment, privacy concerns, or fear
of negative reactions when the researcher is physically present.
2.3.3 Measures
Mindful Attention Awareness Scale (MAAS: Brown & Ryan, 2003). The MAAS consists
of 15 items assessing the absence of a single factor encompassing attention to and
awareness of the present reality in daily life (e.g., “I find myself preoccupied with the
future or the past”), using a 6-point Likert-scale (1 = almost always and 6 = almost
never). This reversed-items measure was supported by the authors’ argument that
for most people, mindless states are more accessible than mindful states. The MAAS
was designed to exclude any attitudinal components (e.g., acceptance, trust,
patience, empathy), motivational components (the “why” of awareness attention),
and other constructs related to well-being. The original study of the MAAS showed
satisfactory reliability for undergraduate students and adult populations ( = .82 and
.87, respectively), as well as convergent and discriminant validity with various
41
constructs of psychological well-being (Brown & Ryan, 2003). Higher MAAS scores
were also related to (i) less reactivity to threatening emotional stimuli as indicated
by bilateral amygdale response and prefrontal cortical activation (Creswell, Way,
Eisenberger, & Lieberman, 2007), and (ii) less self-reported aggression (as measured
by Buss & Perry’s Aggression Questionnaire, 1992) in both undergraduate students
(Heppner et al., 2008), adult populations (Borders et al., 2010), and prospective
criminal justice professionals (Kelly & Lambert, 2012). A recent validation of the
MAAS was provided in MacKillop and Anderson (2007) and Brown et al. (2011). The
MAAS receives the strongest support in a number of studies that explicitly tested the
predictive validity of various mindfulness scales (Sauer et al., 2013).
Brief Self-Control Scale (Brief SCS: Tangney, Baumeister, & Boone, 2004). The SCS was
developed based on extensive review of published studies on self-control processes
and failures. The brief version of the SCS covered the same range of content with the
full 36-item version, i.e., control over thoughts, emotional control, impulse control,
performance regulation, and habit breaking Participants responded to 13 statements
reflecting how they typically are (e.g., “People can count on me to keep on
schedule”) using a 5-point Likert-scale (1 = not at all and 5 = very much). The authors
reported good internal reliability ( = .83 to .85) and test-retest reliability ( = .87),
as well as correlation with numerous adaptive functioning. Higher scores on the brief
SCS were also associated with higher trait of mindfulness (as measured by Brown &
Ryan’s MAAS, 2003) in the context of romantic relationship (Barnes et al., 2007) and
amongst frequent student gamblers (Lakey et al., 2007). A recent meta-analytic
review found that compared to other widely used self-reported measures of self-
control (i.e., the Barratt Impulsiveness Scale [Patton, Standfod, & Barratt, 1995] and
the Low-Self-Control Scale [Grasmick, Tittle, Bursik, & Arneklev, 1993]), the SCS
showed a stronger relationships to overall behaviour and allowed for a more
finegrained analysis across various life domains (de Ridder et al., 2012).
Aggression Questionnaire (AQ: Buss & Perry, 1992). The AQ is one of the most
broadly used self-report measures of aggression. It consists of four subscales, i.e.,
physical aggression (9 items, e.g., “If somebody hits me, I hit back”), verbal
aggression (5 items, e.g., “I tell my friends openly when I disagree with them”),
42
anger (7 items, e.g., “When frustrated, I let my irritation show”), and hostility (8
items, e.g., “I am sometimes eaten up with jealousy”), along with a composite of the
29-score of trait aggression. The first two subscales represent the behavioural or
overt component of aggression, followed by the emotional and the cognitive
components. Participants indicated how accurately each item described the way in
which they act when they feel angry or aggressive using a 5-point Likert-scale (1 =
very inaccurate and 5 = accurate). The authors reported good test-retest reliability (
= .80, .76, .72, .72, .80 for physical aggression, verbal aggression, anger, hostility, and
total score, respectively), and significant correlations with traits such as
emotionality, impulsiveness, and competitiveness, as well as peer reports of
aggression. Importantly, sex differences were found mostly in terms of physical
aggression, followed by verbal aggression and hostility, but not in anger (Buss &
Perry, 1992). Sex differences in aggression (males scoring higher than females in the
behavioural components of aggression) have been consistently shown outside the
U.S. samples, include in British (Archer, Kilpatrick, & Bramwell, 1995), Japanese
(Nakano, 2001), Hungarian (Gerevich, Bácskai, & Czobor, 2007), and Japanese and
Spanish (Ramirez, Andreau, & Fujihara, 2001) samples.
Situational Triggers of Aggressive Responses scale (STAR scale: Lawrence, 2006). The
STAR consists of 22 items compromising two sub-scales, namely sensitivity to
provocations (SP, 12 items, e.g., “A friend betrays me”) and sensitivity to frustrations
(SF, 10 items, e.g., “I experience family dispute”). Each item is rated using a 5-point
Likert-scale (1 = very inaccurate and 5 = very accurate). The author reported good
internal reliability ( = .82 and .80 for SP and SF, respectively) and convergent
validity with measures of traits associated with aggressive behaviour, including with
Buss and Perry’s (1992) AQ. Additionally, the factor congruency for both subscales
was stable across males and females (Lawrence, 2006).
Deliberate Self-Harm Inventory (DSHI: Gratz, 2001). Participants indicated “yes” or
“no” to a list of 17 items (e.g., “Have you ever intentionally (i.e., on purpose), cut
your wrist, arms, or other area(s) of your body? (without intending to kill
yourself)?”). If the answer was yes, participants were asked to rate the number of
times they have administered each act. Responses to item number 17 (i.e., “Have
43
you ever intentionally done anything else to hurt yourself that was not asked about
in this questionnaire? If yes, what did you do to hurt yourself?”) are assessed
qualitatively, and would only be included if they were consistent with the definition
of self-harm used here, i.e., “the deliberate, direct destruction of body tissue without
conscious suicidal intent, but resulting in injury severe enough for tissue damage
(e.g., scarring) to occur” (Gratz, 2001, p.255). High internal consistency ( = .82) and
test-retest reliability ( = .68, p < .01), as well as good construct, convergent, and
discriminant validity were reported (Gratz, 2001). In the current study, we followed
Gratz & Chapman’s (2007) recommendation to use a cutoff of five episodes of self-
harm, since this number is considered to be clinically meaningful. Thus as a final
score, we derived a dichotomous self-harm variable by assigning a score of “1” to
participants who provided the rating of “five times or more” on any items, and a
score of “0” to the rest of the participants. Evidence for the construct and predictive
validity of the DSHI was further reported in Fliege et al. (2006).
Hospital Anxiety and Depression (HADS: Zigmond & Snaith, 1983). Participants rated
14 questions relating to their current state of anxiety (7 items, e.g., “Worrying
thoughts go through my mind”) and depression (7 items, e.g., “I have lost interest in
my appearance”) on scale 0 to 3 (0 = not at all and 3 = very often). A validity study of
the HADS (Bjelland, Dahl, Haug, & Necklemann, 2002) demonstrated its reliability in
assessing the symptom severity and occurrence of anxiety disorders and depression
(mean = .83 for anxiety and .82 for depression) amongst somatic, psychiatric, and
primary care patients and general population. In the normative sample, the total
levels of anxiety and depression from 0 to 7 is considered to be normal, 8 to 10 is
borderline, and 11 to 21 is of clinical significance (Zigmond & Snaith, 1983).
2.3.4 Power calculation
A precise effect size for this study could not be predicted due to of a lack of previous
similar research. To detect a small effect (d = .10) from six predictor variables (i.e.,
trait mindfulness, trait self-control, sensitivity to provocations and frustrations,
anxiety and depression), and given the power of .80 and an alpha level of .05, this
44
study required 143 participants as calculated by G*Power 3 Version 3.1.7 (2013; see
Faul, Erdfelder, Buchner, & Lang, 2009). If the demographic factors (i.e., current
educational status, sex, age, race/ethnicity, and current mindfulness practice) are
included as predictors, the total sample required would be 179.
2.3.5 Procedure
Since the study aimed to access both university and non-university students,
potential participants were contacted not only via posters/leaflets on campus and
the University of Nottingham mailing lists, but also through the social networking
site (Facebook) by asking the researchers’ colleagues from non-academic groups to
post the advertisement of the study on their Facebook pages. A statement about the
confidentiality nature of the study was made in these advertising materials
(Appendix 2.2), along with its direct website link.
Interested participants were first shown an electronic information sheet indicating
the purpose of the study and the contact details of the researchers (Appendix 2.3).
They were told that the study was not concerned with their actual levels of
mindfulness, self-control, aggression, and self-harm—but in how those behaviours
and propensities were associated with each other. An electronic consent form was
then displayed with relevant information (Appendix 2.3), including a statement that
participants are free to withdraw at any time. Those who chose to participate were
then presented with a series of questions (Appendix 2.1). The MAAS was presented
first, followed by the SCS, HADS, STAR, AQ, and DSHI, respectively. In any case,
starting a survey with questions that are not particularly personal or sensitive is a
good strategy to minimise unwillingness to continue the study (Goodwin, 2010).
Upon completion of the study, participants were required to provide demographic
information consisting of current educational status (university/non university
student), sex, age, race/ethnicity, and current mindfulness practice (defined as “a
specific meditational practice in directing attention without making any judgment”).
They were also given an option to enter an email address to win a £25 prize draw
45
incentivising the study. At the end of the survey, participants were debriefed and
given help and support information (Appendix 2.5).
2.3.6 Data analysis
Analyses were performed on the data using the IBM SPSS statistics 20.0 (2011). The
relationships between demographic factors and self-reported measures were
assessed using one-way ANOVAs or independent t-tests (for current educational
status, sex, race/ethnicity, and current mindfulness practice), Pearson’s correlation
(for age), and Chi-square test (for the dichotomous self-harm measure).
2.3.6.1 Link between mindfulness, self-control, and aggression
The associations amongst self-reported measures were explored using zero-order
correlations. On examination of these correlations, Hypothesis 1 would be
confirmed if there is a positive correlation between trait mindfulness and trait self-
control and negative correlations between these traits and trait aggression.
The possible relationships between mindfulness, self-control, and aggression were
tested with bootstrapping method. Specifically, we used (i) Preacher and A. F.
Hayes’s (2011) macro called INDIRECT (for Hypothesis 2 and 3), and (ii) A. F. Hayes’s
(2012a) macro called PROCESS (for Hypothesis 4).
For Hypothesis 2 and 3, we tested whether the relationship between trait
mindfulness (predictor) and aggression (outcome) would be mediated by trait self-
control (proposed mediator). The main outcome was trait aggression. Additionally,
the four aggression subscales (physical and verbal aggression, anger, and hostility)
were analysed separately as outcomes. A mediation model with only one mediating
variable is known as simple mediation (see Figure 2.1).
Traditionally, mediation testing followed a four-step causal approach popularised by
R. M. Baron and Kenny (1986; see also Kenny, 2013). With this approach, several
hierarchical regression analyses are conducted and significance of the coefficients is
46
examined at each step. At Step 1, the outcome (Y) is regressed on the predictor (X)
to determine a significant total effect (c path). At Step 2, the proposed mediator (M)
is regressed on the predictor (a path). At Step 3, the outcome is regressed on both
the mediator and the predictor. As such, the predictor is controlled to establish the
effect of the mediator on the outcome (b path). One would proceed to Step 4 if
there are significant relationships from Step 1 to 3. However, current researchers
have strongly advocated that it is not always necessarily for the c path to be
significant in order to establish mediation (for a review, see e.g., MacKinnon,
Fairchild, & Fritz, 2007; Zhao, Lynch & Chen, 2010). At Step 4, if the direct effect of
the predictor on the outcome is zero after controlling for the mediator (c’ path) then
the finding supports full mediation. If the direct effect decreases but the reduction is
still different from zero then the finding supports partial mediation. It should be
noted that statistically, Step 3 and Step 4 are estimated in the same equation. The
amount of mediation is called the indirect effect (ab path).
Figure 2.1. Simple mediation model. The causal effect of X on Y (c) is apportioned through itsindirect effect on Y through M (ab) and its direct effect on Y (c’). X = predictor (trait
mindfulness), Y = outcome (trait aggression or self-harm), M = proposed mediator (trait self-control). a, b, c, c’ = unstandardised regression coefficient. Reprinted from Kenny (2013).
In the causal approach, the size of the mediation is not directly estimated, but is
mathematically derived as the product of the a and b paths. Since the total effect of
the predictor on the outcome is equal to the sum of the direct and indirect effects, c
47
= c’+ ab, the indirect effect is calculated as ab = c – c’. The main problem with this
calculation is that a confidence interval for the population indirect effect could not
be obtained (Pituch, Whittaker, & Stapleton, 2005; for a thorough discussion on the
limitations of the causal approach consult MacKinnon, Lockwood, Hoffman, West, &
Sheets, 2002; Preacher & A. F. Hayes, 2008). An initial attempt to perform a single
test of ab was proposed by Sobel (1982). The Sobel test, however, assumes
normality of the sampling distribution of the indirect effect, which is typically
violated in practice (e.g., MacKinnon, Lockwood & Williams, 2004; Preacher & A. F.
Hayes, 2004; Shrout & Bolger, 2002).
Conversely, with bootstrapping method, the sampling distribution of ab is estimated
empirically; therefore no assumptions are made about this distribution (see e.g.,
MacKinnon et al., 2004; Preacher & A. F. Hayes, 2008). To bootstrap an indirect
effect, the available sample size n is re-sampled with replacement intensively for a
total of k times to estimate the a and b as usual (Preacher & A. F. Hayes, 2008,
suggest using 5,000 resamples for final reporting). These estimates of a and b are
used to calculate ab* (the indirect effect in a single resample), and the distribution of
the k values of ab* provides a nonparametric approximation of the sampling
distribution of ab. The mean of the k estimates of ab* represents the indirect effect.
Since the mean of the k is not exactly equal with the indirect effect, a correction for
bias is made, typically using confidence intervals (MacKinnon et al., 2004; Preacher &
A. F. Hayes, 2008). If the bias-corrected confidence intervals does not include zero,
one can be confident that a significant mediation has occurred.
Hypothesis 2 would be supported if there is a zero in the bias-corrected confidence
intervals of path ab (indirect effect of self-control on the link between mindfulness
and aggression). Hypothesis 3 would be confirmed if path c’ (the direct effect of
mindfulness on aggression) is still significant after the inclusion of trait self-control.
For Hypothesis 4, we tested whether the relationship between trait self-control
(predictor) and aggression (outcome) would be moderated by trait mindfulness
(proposed moderator). Trait aggression was used again as the main outcome, and
the four aggression subscales were analysed separately as outcomes. A moderator
48
model with only one moderator variable is known as simple moderation (see Figure
2.2). The variables are mean-centered so that their coefficients are interpretable
within the range of the data (A. F. Hayes, 2012b).
Figure 2.2. Conceptual and statistical models for simple moderation. In the statisticalmodel, Y is estimated as a weighted function of X, M, and the product of X and M (XM).X = predictor (trait self-control), Y = outcome (trait aggression or self-harm), M =moderator (trait mindfulness). c1, c2, c3 = unstandardised regression coefficient.Reprinted from A. F. Hayes (2012b, p.33).
Theoretically, X is depicted to exert an influence on Y, and this effect is proposed to
be influenced or moderated by M (the Conceptual Model in Figure 2.2). These effects
are estimated mathematically (the Statistical Model in Figure 2.2, Statistical Model)
in the form of a linier equation: Y = i + c1X + c2M + c3XM + ey. From this equation, it
can be seen that the effect of X on Y is a function of M, since Y = i + (c1 + c3M)X + c2M
+ ey (for further details, consult A. F. Hayes, 2012b). The (c1 + c3M) function
represents the conditional effect of X on Y or “simple slope” for X, in which c1
estimates the effect of X on Y when M = 0, and c3 estimates how much the effect of X
on Y changes as M changes by one unit. The main focus in a moderation model is
whether c3, the interaction coefficient between X and M, is statistically different
from zero. Additionally, the significance of the change in the total variance in Y due
to this interaction is also useful. When the null hypothesis is rejected, the magnitude
of the conditional effect of X at various values of M is elaborated, along with a
standard error, t, and p-value. For dichotomous M, this effect is derived at each of
the two values of M. If M is continuous, M can be set to various values that
49
represent low (a standard deviation below the mean), moderate (the mean), and
high (a standard deviation above the mean).
Hypothesis 4 would be confirmed if c3, the interaction coefficient between self-
control and mindfulness, is statistically different from zero.
2.3.6.2 Link between mindfulness, self-control, aggression, and self-harm
Using independent sample t-test, we tested the differences between self-harmers
and non-self harmers on the measures of (i) trait aggression and its four subscales
(Hypothesis 5), and (ii) trait mindfulness and self-control (Hypothesis 6).
For Hypothesis 7, 8a, and 8b, the same INDIRECT macro (Preacher & A. F. Hayes,
2011) was used with the dichotomous self-harm variable (yes vs. no) as the
outcome, to test the effect of the predictor (trait mindfulness) and mediator (trait
self-control) of self-harm.
Principally, when the outcome is dichotomous, the logit-transformed probability is
modeled as a linear relationship with the predictors, known as logistic regression
analysis (see e.g., MacKinnon & Dwyer, 1993). INDIRECT macro estimates the b, c,
and c’ paths accordingly using logistic regression, based on a Netwon-Raphson
iteration algorithm. To determine if the overall model is significant, the statistics for
maximumlikelihood estimate, McFadden R2, Cox & Snell R2, Nagelkerke R2 are
produced. The importance of each variable is indicated by the significant p value of
Wald statistic. Crucially, the confidence intervals for the indirect effect (ab) are still
estimated as the product of the c’ and the b paths. However since ab and c are
scaled differently, the equation c – c’ cannot be used as a substitute for the total
indirect effect or as the proportion of the effect that is mediated (Preacher & A. F.
Hayes, 2008). Hence the bias-corrected confidence intervals of path ab would
provide information about the mediation role of self-control on the link between
mindfulness and self-harm (Hypothesis 7). Hypothesis 8a and 8b would be confirmed
if path c’ (the direct effect of mindfulness on self-harm) is still significant after the
inclusion of trait self-control (and controlling for anxiety and depression).
50
The moderation model of trait mindfulness on the link between trait self-control and
self-harm (Hypothesis 9) would be tested in the same way with Hypothesis 4.
2.3.6.3 Role of individual differences in SP and SF
The role of SP and SF on aggression and self-harm (Hypothesis 10) could be
supported if (i) the zero-order correlations indicate a positive correlation between SP
and SF and trait aggression (ii) independent sample t-test reveal the differences
between self-harmers and non-self harmers on the measures of SP and SF.
The moderation of mindfulness and self-control on the link between SP and SF and
aggression (Hypothesis 11a) and self-harm (Hypothesis 11b) would be tested with
the same PROCESS macro (A. F. Hayes, 2012a). On separate analyses for SP and SF,
we used trait mindfulness and self-control as the moderators; trait aggression, the
four aggression subscales, and self-harm as the outcomes. The moderator model in
Figure 2.2 is modified to include two moderators (see Figure 2.3).
Figure 2.3. Conceptual and statistical models for additive moderation . X = predictor (SP orSF), Y = outcome (aggression or self-harm), M = first moderator (trait mindfulness), W =
second moderator (trait self-control). c1, c2, c3, c4, c5 = unstandardised regression coefficient.Reprinted from A. F. Hayes (2012b, p.33).
The effect of X on Y is proposed to be moderated by both M and W. Mathematically,
Y = i + c1X + c2M + c3W + c4XM + c5WM + ey. It can be seen that the effect of X on Y is
additively dependent on both M and W, since Y = i + (c1 + c4M + c5W)X + c2M + c3W +
51
ey. The conditional effect of X on Y is = (c1 + c4M + c5W). If both c4 and c5 are
statistically different from zero, the magnitude of the conditional effect is being
inspected at various combinations of M and W.
2.3.7 Data preparation
Prior to further analysis, we examined all measures for extensive missing data
leaving in total 241 participants. The pattern of the missing data was: 25 subjects did
not filled out any items on the survey, 11 did not continue to the second half of the
first survey (the MAAS), 1 did not continue to the second survey (the SCS), 3 did not
continue to the third survey (the HADS), 10 did not continue to the fourth survey
(the STAR), 6 did not continue to the fifth survey (the AQ), 11 did not continue to the
second half of the fifth survey, and 1 did not continue to the last survey (the DSHI).
These 68 participants were removed from the analysis.
2.4 RESULTS
2.4.1 Influence of demographic factors
We firstly conducted analyses on the associations between demographic factors (i.e.,
current educational status, sex, age, race/ethnicity, current mindfulness practice)
and our predictors (trait mindfulness, trait self-control, sensitivity to provocations
[SP] and sensitivity to provocations [SF]). Any significant demographic factors would
be controlled in the relevant subsequent analyses.
Age was positively associated with mindfulness (r = .21, p < .01) and self-control (r =
.25, p < .001). Non-students participants were higher in mindfulness (t(239) = 4.18, p
< .0001; M non-students = 4.38, SD = .68 vs. M students = 3.77, SD = .76) and self-
control (t(239) = 4.45, p < .0001; M non-students = 3.47, SD = .58 vs. M students =
2.95, SD = .60), but lower in SP (t(239) = 1.99, p < .05; M non-students = 2.69, SD =.78
vs. M students = 2.99, SD = .76). SF was neither related to age (r = .04, p = .46) nor
educational status (t(239) = -.46, p = .15).
52
As the non-students participants were older than students (t(234) = 9.41, p < .0001;
M non-students = 32.10, SD = 4.76 vs. M students = 22.68, SD = 5.17), a partial
correlation was carried out between age, mindfulness, and self-control after
controlling for educational status. Although age was still related to self-control (r =
.14, p < .05), it was no longer related to mindfulness (r = .08, p = .20). Hence, age and
current educational status were both included as covariates in the main analyses.
Sex differences were not linked to mindfulness (t(235) = .70, p = .48), self-control
(t(235) = .57, p = .57), SP (t(199.76) = .08, p = .93), or SF (t(235) = -1.54, p = .13). No
differences associated with ethnicity were found in mindfulness (F(2,231) = .81, p =
.44), self-control (F(2,231) = 1.63, p = .20), SP (F(2,231) =.81, p = .45), or SF (F(2,231)
= .56, p = .57). Interestingly, participants with mindfulness experience did not
indicate differences in mindfulness (F(3,237) = .94, p = .42), self-control (F(3,237) =
.65, p = .56), SP (F(3,237) = 1.66, p = .18), or SF (F(3,237) = 1.39, p = .25).
2.4.2 Trait aggression scores
A mean score of aggression (i.e., the average score across four AQ subscales) was
calculated as a main trait aggression measure. Sex differences in aggression are
shown in Table 1 (N = 237 participants, as 4 subjects did not report their sex). No sex
differences were found in general trait aggression, anger, or hostility. Males,
however, scored significantly higher than females on physical and verbal aggression.
The results support the notion that the effect of sex differences is more salient with
regard to aggressive behaviours than with aggressive feelings (Lawrence, 2006).
Table 2.1.Means for aggression measures based on participants’ sex
Aggression measuresMales (n = 85) Females (n = 152)
tM (SD) M (SD)
Trait aggression 2.62 (0.62) 2.47 (0.67) t(235) = 1.67, p = .10Physical Aggression 2.48 (0.87) 2.03 (0.80) t(235) = 4.01***
Verbal Aggression 3.03 (0.87) 2.78 (0.80) t(235) = 2.26*
Anger 2.34 (0.71) 2.51 (0.84) t(198.48) = -1.64, p = .10Hostility 2.63 (0.78) 2.57 (0.86) t(235) = .51, p = .61*p < .05;
**p < .01;
***p < .001
53
Higher levels of trait aggression were also linked to ethnicity (F(2,231) = 4.85, p < .01;
M Asian = 2.75, SD = .55 vs. M “Others” = 2.53, SD = .58 vs. M White = 2.44, SD =
.68). In addition, Asian participants reported more anger than White (F(2,231) = 6.26,
p < .01; M Asian = 2.76, SD = .68 vs. M White = 2.35, SD = .81). Participants who
reported engaging in current mindfulness practice were higher in trait aggression
(F(3,237) = 3.92, p < .01), anger (F(3,237) = 4.51, p < .01), and hostility (F(3,237) =
3.02, p < .05), compared with those with no experiences. Neither age nor current
educational status was related to the aggression measures.
2.4.3 Self-harm scores
A score of “1” was assigned to those who reported having engaged in self-harm five
or more times in any of the 17 DSHI items (Gratz & Chapman, 2007). For item
number 17 (i.e., qualitative description of self-harm methods that had not been
included in the first 16 items), we reassessed participants’ responses against Gratz’s
(2001) definition of self harm. Descriptions like “Hair straightens to burn myself” was
included, but “Refusing food, refusing sleep, smoking, drinking, deliberate vomiting”
was excluded. In total, 49 participants (20.30%) were qualified as self-harmers.
With reference to the methods of self-harm, the most frequent method was skin
cutting (used by 22% of the total sample, with equally prevalent across males and
females [2(1) = 1.86, p = .17; females = 11.30%, males = 5.90%]), resembling Gratz’s
(2001) developmental study of the DSHI.
The only demographic factor that affected self-harmful behaviours was age
(t(118.25) = 3.64; M self-harmers = 21.76, SD = 4.00 vs. M non self-harmers = 24.43,
SD = 6.32). We dichotomised ethnicity to White and non-White because the
expected count for the “Other” ethnic background was less than 5. Self-harm status
did not vary as a function of ethnicity (2(1) = 2.58, p = .11), current mindfulness
practice (2(1) = .06, p = .80), or participant’s sex (2(1) = .74, p = .39; n females = 34
[22.40%] vs. n males = 15 [17.60%]). Although classic self-harm literature typically
demonstrates that females engage in more self-harm behaviours, sex differences in
54
self-harm may be more pronounce in early adolescence (for a review, see Hamza et
al., 2012). In this sample, the odds of females engaging in self-harmful behaviours
were only 1.38 times higher than males.
2.4.4 Preliminary analysis
Table 2.2 (below diagonal) presents the zero-order correlations amongst continuous
variables and descriptive statistics (N = 241). Internal reliabilities of measures ranged
from adequate ( = .74) to quite high ( = .90). All measures were generally
associated with each other, and in the expected direction. Specifically, mindfulness
and self-control were positively correlated to each other and negatively correlated to
all of the aggression variables. Hypothesis 1 was confirmed such that individuals who
were more mindful and more self-controlled were less aggressive.
Table 2.2.Zero order correlations and psychometric properties of measures
Measures 1 2 3 4 5 6 7 8 9
MAAS (1) 1.00SCS (2) .50** 1.00Total AQ (3) -.36** -.35** 1.00AQ Physical (4) -.20** -.24** .78** 1.00AQ Verbal (5) -.13* -.15* .76** .44** 1.00AQ Anger (6) -.33** -.28** .83** .57** .52** 1.00AQ Hostility (7) -.47** -.44** .78** .45** .44** .55** 1.00STAR Provocations (8) -.18** -.28** .45** .33** .29** .38** .44** 1.00STAR Frustrations (9) -.27** -.34** .46** .28** .26** .45** .46** .74** 1.00
M 3.84 3.02 2.53 2.20 2.88 2.46 2.60 2.95 2.71SD .78 .62 .65 .85 .83 .80 .84 .77 .82
Cronbach .88 .82 .90 .84 .74 78 .80 .88 .87Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; STAR = Situational Triggersof Aggressive Responses; AQ = Aggression Questionnaire*p < .05;
**p < .01;
***p < .001
Additionally, none of these associations were significantly altered when we run
partial correlations controlling for sex, ethnicity, or current educational status. The
relation between trait mindfulness and verbal aggression fell just short of
significance when we control for age (r = -.13, p = .06) and current mindfulness
practice (r = -.12, p = .06).
55
2.4.5 Mediation of self-control on the link between mindfulness and aggression
For Hypothesis 2 and 3, a bootstrapping method with 95% bias-corrected confidence
intervals (based on 5,000 bootstrap resamples, N = 241) was used to test the
mediating role of self-control on the link between mindfulness and aggression.
2.4.5.1 Trait aggression
Hypothesis 2 was confirmed such that self-control significantly mediated the
relationship between mindfulness and trait aggression. Specifically, mindfulness
predicted self-control (B = .40, SE = .04, p < .0001), while trait aggression was
predicted by mindfulness (B = -.30, SE = .05, p < .0001) and self-control (B = -.24, SE =
.07, p < .05). Thus the paths from mindfulness to self-control (a path) and from self-
control to aggression (b path) were both significant. Crucially, self-control mediated
the relationship between mindfulness and trait aggression (ab path: B = -.09, SE =
.03, 95% CI [-.16, -.04]). The direct effect of mindfulness was still significant (B = -.20,
SE = .05, p < .001) when accounting for self-control (c’ path), indicating a partial
mediation. This latter finding supported that mindfulness accounted for significance
variance in trait aggression after accounting for self-control (Hypothesis 3).
A very similar pattern of results emerged when the hypothesised covariates (i.e.,
current educational status, sex, age, race/ethnicity, and current mindfulness
practice) were controlled. Due to some unreported demographic information, the
bootstrapping procedure allowed data from 233 samples only (listwise deletion). All
paths were still significant, and self-control partially mediated the link between
mindfulness and trait aggression (B = -.10, SE = .03, 95% CI [-.17, -.05]). Again,
mindfulness accounted for significant variance in trait aggression after controlling for
covariates (B = -.20, SE = .06, p < .001). Significant partial effects of covariates were
found across sexes (B = -.18, SE = .08, p < .05) and ethnicity (B = .18, SE = .07, p < .05).
On examination of the specific factors, higher levels of trait aggression were
associated with being male and Asian.
56
2.4.5.2 Aggression subscales
To further examine the proposed mediation model, the analysis was repeated for
each of the four trait aggression subscales (see Table 2.3). It can be seen that the
total effect of mindfulness was significant for all aggression subscales (c path). The
effect of self-control was not significant on verbal aggression (B = -.15, p = .60),
therefore no mediation model could proposed on this subscale. A significant
mediation effect was demonstrated for the three other subscales (see ab path). Thus
Hypothesis 2 was partly confirmed for the mediating role of self-control on the
relationship between mindfulness and physical aggression, anger, and hostility.
Further, with the inclusion of self-control, the predicting role of mindfulness on
physical aggression was no longer significant (B dropped from -.22 to -.12).
Mindfulness still, however, significantly predicted anger and hostility, indicating its
unique contribution on these subscales (Hypothesis 3).
Table 2.3.Mediation models with bootstrapping method, using mindfulness as predictor, self-control asmediator, aggression subscales as outcomes
Model testedPhysical
aggressionVerbal
aggressionAnger Hostility
B SE B SE B SE B SE
Mindfulness to aggression(c path)
-.22** .07 -.14* .07 -.34*** .06 -.51*** .06
Mindfulness to self-control(a path)
.40*** .04 .40*** .04 .40*** .04 .40*** .04
Self-control to aggression(b path)
-.26* .10 -.15 .10 -.20* .09 -.37*** .08
Mindfulness to aggression,via self-control (c’ path)
-.12 .08 -.08 .08 -.26*** .07 -.36*** .07
Indirect path (ab path) -.10 a .04 -.06 .05 -.08 a .04 -.14 a .04Bias-corrected bootstrap95% confidence intervals
[-.19, -.03] [-.15, .03] [-.16, -.01] [-.22, -.08]
Note. B = unstandardised regression coefficient; SE = standard error.a
Zero is not included in the 95% confidence interval, indicating that the indirect path is significantlydifferent from zero.*p < .05;
**p < .01;
***p < .001
When the analysis was repeated including the covariates (current educational status,
sex, age, race/ethnicity, and current mindfulness practice), again no mediation
model could proposed on verbal aggression (b path: B = -.15, p = .12). Self-control
57
continued to mediate the link between mindfulness and physical aggression (B = -
.11, SE = .04, 95% CI [-.19, -.04]), hostility (B = -.09, SE = .03, 95% CI [-.16, .03]), and
anger (B = -.09, SE = .04, 95% CI [-.18, -.03]). After the inclusion of self-control,
mindfulness no longer predicted physical aggression (B dropped from -.14 to -.10, p =
.20), but still predicted hostility (B dropped from -.41 to -.36, p < .0001) and anger (B
barely changed from -.25, p < .001). These findings highlighted the mediating effect
of self-control (on physical aggression, anger, and hostility) as well as the unique role
of mindfulness (on anger and hostility) without controlling the covariates.
For clarity of presentation, the partial effects of covariates are presented separately
(Table 2.4). Males scored higher in overt aggression, whereas Asian participants
showed higher physical aggression and hostility. None of the hypothesised
covariates influenced the levels of anger.
Table 2.4.Partial effect of control variables for mediation models with bootstrapping method, usingmindfulness as predictor, self-control as mediator, aggression subscales as outcomes
CovariatesPhysical
aggressionVerbal
aggressionAnger Hostility
B SE B SE B SE B SE
Sex -.49*** .11 -.26* .11 .13 .10 -.11 .09Age -.01 .01 -.01 .01 .01 .01 -.01 .01Ethnicity .20* .10 .17 .10 .17 .09 .19* .09Current educational status -.18 .19 -.13 .19 -.05 .17 -.24 .17Current mindfulness practice .07 .08 .10 .08 .12 .08 .05 .07Note. B = unstandardised regression coefficient; SE = standard errorN = 233 (listwise deletion)*p < .05;
**p < .01;
***p < .001
2.4.6 Moderation of mindfulness on the link between self-control and aggression
To examine the moderating effect of mindfulness on the link between self-control
and aggression (Hypothesis 4), a bootstrapping method (model number 1 in the
“Process” macro for SPSS), with 95% bias-corrected confidence intervals was carried
out. All predictors and proposed moderators were mean-centered.
58
2.4.6.1 Trait aggression
Significant main effects of mindfulness (c2: B = -.20, SE = .06, p < .0001) and self-
control (c1: B = -.24, SE = .07, p < .01) were found on general trait aggression.
However, the interaction effect between mindfulness and self-control was not
significant (c3: B = -.06, SE = .07, p = .46), with insignificant change in the total
variance in aggression (R2 = .20%). Thus Hypothesis 4 was not supported such that
self-controlled individuals who were at the same time also mindful were no less
aggressive then self-controlled individuals who were not mindful. Levels of general
aggression were predicted separately by trait mindfulness and self-control.
2.4.6.2 Aggression subscales
The moderation analysis was repeated for each of the four trait aggression subscales
(Table 2.5). Overall, mindfulness did not moderate any relationships between self-
control and the aggression subscales. Individuals who were more self-controlled
were less physically aggressive and less hostile regardless their levels of mindfulness,
whereas individuals who were more mindfulness showed less hostility and anger
regardless of their levels of self-control.
Table 2.5.Moderation models with bootstrapping method, using self-control as predictor, mindfulnessas moderator, aggression subscales as outcomes
Model testedPhysical
aggressionVerbal
aggressionAnger Hostility
B SE B SE B SE B SE
Self-control (c1) -.26* .10 -.15 .10 -.19 .09 -.37*** .09Mindfulness (c2) -.12 .08 -.08 .08 -.26*** .07 -.36*** .07Self-control xMindfulness (c3)
-.05 .64 -.12 .10 -.04 .09 -.02 .09
R2 on interaction .08%, p = .65 .56%, p = .24 .05%, p = .70 .01%, p = 83Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
59
2.4.7 Link between self-harm and harm to others
Hypothesis 5 was tested by comparing self-harmers vs. non self-harmers in terms of
their aggression levels. Those who self-harmed showed higher levels of general trait
aggression (t(239) = 2.78, p < .01; M self-harmers = 2.76, SD = .64 vs. M non self-
harmers = 2.47, SD = .64). In addition, differences between self-harmers and non-
harmers were found for physical aggression (t(239) = 3.58, p < .001; M self-harmers =
2.58, SD = .95 vs. M non self-harmers = 2.10, SD = .80) and anger (t(239) = 2.38, p
<.05; M self-harmers = 2.70, SD = .84 vs. M non self-harmers = 2.40, SD = .78). No
differences were found in verbal aggression (t(97) = 1.57, p = .12; M self-harmers =
3.05, SD = .78 vs. M non self-harmers = 2.84, SD = .84) or hostility (t(97) = 1.20, p =
.23; M self-harmers = 2.72, SD = .93 vs. M non self-harmers = 2.56, SD = .81).
2.4.8 Link between mindfulness, self-control, and self-harm
Provided that to some extent self-harm was positively related to aggression, we
expected that the association between self-harm and mindfulness and self-control
would mirror the association between these predictors and trait aggression
(Hypothesis 6). As expected, self-harmers (n = 49) scored lower than non self-
harmers (n = 192) in both trait mindfulness (t(239) = 3.58, p <.001; M self-harmers =
3.62, SD = .89 vs. M non self-harmers = 3.90, SD = .74) and self-control (t(239) = 2.38,
p < .05; M self-harmers = 2.77, SD = .54 vs. M non self-harmers = 3.08, SD = .62).
Additionally, we assessed the role of anxiety (M = 1.17, SD = .58) and depression (M
= .63, SD = .50) on self-harm. High internal validity was shown on both scales ( = .81
for anxiety, and .79 for depression), and they were strongly related to each other (r =
.58, p < .01). None of the participants reported levels of anxiety and depression
beyond normal (M of total anxiety and depression range from 0 to 5.43). Self-
harmers were significantly more anxious (t(239) = 3.27, p < .001; M self-harmers =
1.44, SD = .67 vs. M non self-harmers = 1.10, SD = .54). They did not report more
depression than non self-harmers (t(61.68) = 1.76, p = .08; M self-harmers = .76, SD =
.63 vs. M non self-harmers = .59, SD = .46).
60
2.4.9 Mediation of self-control on the link between mindfulness and self-harm
To further examine if the mediation model of self-control also applied on the link
between mindfulness and self-harm, we carried out another bootstrapping method
with the dichotomous self-harm variable (yes vs. no) as dependent variable. In
previous analysis (see section 2.4.5.1), it was shown that mindfulness predicted self-
control (B = .39, SE = .04, p < .0001). Results for 95% bias corrected bootstrap
confidence intervals confirmed the mediating role of self-control on the link between
mindfulness and self-harm. Specifically, self-harm was predicted by mindfulness (B =
-.47, SE = .21, Wald = 4.90, p < .05) and self-control (B = -.74, SE = .31, Wald = 5.61, p
< .05). Self-control mediated the link between mindfulness and self-harm (B = -.30,
SE = .13, 95% CI [-.58, -.05]), thus providing support for Hypothesis 7. The model
explained between 4.4% (McFadden R2 and Cox & Snell R2) and 6.9% (Nagelkerke R2)
of the variance in self-harm status. Even though there was not enough evidence for a
full mediation (i.e., regression coefficient was not reduced to zero), the direct effect
of mindfulness was no longer significant (B = -.20, SE = .20, Wald = .65, p = .42) when
accounting for self-control. Thus Hypothesis 8a could not be supported because
mindfulness did not account for significance variance in self-harm over self-control.
The mediation model of self-control became non-significant (B = -.17, SE = .11, 95%
CI [-.40, .03]) when we controlled for anxiety and depression. Prediction in explaining
the variance in self-harm status increased between 6.61% (Cox & Snell R2) or 6.77%
(McFadden R2) and 10.39% (Nagelkerke R2), in which self-harm was predicted by
anxiety (B = .83, SE = .39, Wald = 4.64, p < .05), but not by depression (B = -.11, SE =
.40, Wald = .07, p = .79). Again, mindfulness did not account for significant variance
of self-harm over self-control and anxiety and depression (B = .006, SE = .26, Wald =
.0005, p = .98), thus no support was found for Hypothesis 8b.
When the analysis was repeated including the covariates (i.e., current educational
status, sex, age, race/ethnicity, and current mindfulness practice; N = 233), the
predicting role of anxiety on self-harm fell just short of significance (B = .77, SE = .40,
Wald = 3.61, p = .06). Older participants were less likely to self-harmed (B = -.09, SE =
.05, Wald = 3.89, p < .05).
61
2.4.10 Moderation of mindfulness on the link between self-control and self-harm
Given that mindfulness did not provide incremental validity over self-control in
predicting self-harm, it would be unlikely that mindfulness would act as a moderator
between self-control and self-harm (Hypothesis 9). When we conducted a
bootstrapping method with 95% bias-corrected confidence intervals, the moderation
model was not significant, explaining only 4.90% (Cox & Snell R2) or 4.97%
(McFadden R2) and 7.70% (Nagelkerke R2) variance in self-harm status. Self-harm
was only predicted by self-control (c1: B = -.78, SE = .31, p < .05). There were no
significant main effect of mindfulness (c2: B = -.27, SE = .25, p = .28), and interaction
effect between mindfulness and self-control (c3: B = -.43, SE = .39, p = .25). Thus
Hypothesis 9 could not be supported. For self-controlled individuals, self-harm levels
were the same regardless of mindfulness.
2.4.11 Link between SP and SF and aggression/self-harm
From the zero-order correlations (see Table 2.1), it can be seen that both sensitivity
to provocations (SP) and frustrations (SF) were positively correlated to general trait
aggression and the four aggression subscales (p < .01). Differences between self-
harmers and non self-harmers were also found in SF (t(239) = 2.13, p < .05; M self-
harmers = 2.93, SD = .76 vs. M non self-harmers = 2.65, SD = .83), but not SP
(t(95.25) = 1.50, p = .14; M self-harmers = 3.08, SD = .61 vs. M non self-harmers =
2.92, SD = .80). These findings provided a partial support for Hypothesis 10, such that
individual differences in SP and SF were positively associated with trait aggression,
but only SF was related to self-harm.
2.4.12 Moderation of mindfulness and self-control on the link between SP and SF
and aggression/self-harm
The moderating role of mindfulness and self-control on the link between SP and SF
and aggression (Hypothesis 11a) and self-harm (Hypothesis 11b) were tested using a
62
bootstrapping method with 95% bias-corrected confidence intervals for additive
moderation (model “number 2” in the PROCESS macro). All predictors and proposed
moderators were mean-centered.
2.4.12.1 Trait aggression
No significant interaction effects were found between SP and self-control (c4: B = -
.09, SE = .08, p = .27, R2 = .36%) or between SP and mindfulness (c5: B = -.07, SE = .05,
p = .22, R2 = .44%) on general trait of aggression. There was a small but non-
significant change in the total variance in aggression due to the inclusion of both
moderators (R2 = 1.48%, p = .08). Trait aggression was predicted separately by SP (c1:
B = .34, SE = .05, p < .0001), self-control (c2: B = -.14, SE = .04, p < .05), and
mindfulness (c3: B = -.19, SE = .05, p < .001).
A very similar pattern of results occurred for sensitivity to frustrations (SF). No
significant interaction effects emerged between SF and self-control (c4: B = .06, SE =
.08, p = .46, R2 = .17%), or between SF and mindfulness (c5: B = -.10, SE = .06, p = .09,
R2 = .89%). The change in the total variance in aggression due to the inclusion of
both moderators was non-significant (R2 = .89%, p = .23). Trait aggression was again
predicted separately by SF (c1: B = .30, SE = .05, p < .0001), self-control (c2: B = -.14,
SE = .07, p < .05), and mindfulness (c3: B = -.15, SE = .06, p < .01).
Altogether, Hypothesis 11a could not be supported. Individuals who were higher in
SP and SF but at the same time also mindful or self-controlled showed no less
general propensity to behave aggressively than those who were also higher SP and
SF individuals but were not mindful or self-controlled.
2.4.12.2 Aggression subscales
The moderation analysis was repeated for aggression subscales (Table 2.6 and 2.7).
The moderation of mindfulness on the link between SP and anger (B = -.15, SE = .08,
p < .05, R2 = 1.26%), and SF and hostility (B = .06, SE = .08, p = .46, R2 = .17%) was
63
confirmed. Specifically, individuals who were more sensitive to provocations but at
the same time were mindful reported less anger. Those who were more sensitive to
frustrations but at the same time were mindful were less hostile.
Table 2.6.Moderation models with bootstrapping method, using sensitivity to provocations (SP) aspredictor, mindfulness and self-control as moderators, and aggression subscales as outcomes
Model testedPhysical
aggressionVerbal
aggressionAnger Hostility
B (SE); R2 B (SE); R2 B (SE); R2 B (SE); R2
Self-control (c2) -.16 (.10) -.05 (.10) -.11 (.09) -.25** (.08)
SP (c1) .32*** (.07) .30*** (.07) .34*** (.06) .38*** (.06)
SP x Self-control (c4)-.08 (.12);
.17%-.13 (.12);
.44%-.03 (.10);
.03%-.12 (.10);
.38%Mindfulness (c3) -.10 (.08) -.07 (.08) -.23*** (.07) -.34*** (.07)SP x Mindfulness
(c5)
-.04 (.08);.07%
-.002 (.08);.00%
-.15 (.08)*;1.26%
-.10 (.07);.50%
R2 on bothinteractions
.44%, p = .55 .57%, p = .48 1.88%, p = .06 1.63%*
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
Table 2.7.Moderation models with bootstrapping method, using sensitivity to frustrations (SF) aspredictor, mindfulness and self-control as moderators, and aggression subscales as outcomes
Model testedPhysical
aggressionVerbal
aggressionAnger Hostility
B (SE); R2 B (SE); R2 B (SE); R2 B (SE); R2
Self-control (c2) -.18 (.10) -.07 (.10) -.07 (.09) -.26** (.08)
SF (c1) .23*** (.06) .24*** (.07) .37*** (.06) .34*** (.06)
SF x Self-control (c4)-.01 (.12);
.00%-.09 (.12);
.24%-.05 (.10);
.06%-.10 (.10);
.38%Mindfulness (c3) -.08 (.08) -.04 (.08) -.19** (.07) -.30*** (.07)
SF x Mindfulness (c5)-.08 (.08);
.38%-.05 (.08);
.15%-.12 (.07);
.89%-.14* (.07);
1.07%R2 on bothinteractions
.45%, p = .55 .28%, p = .70 .92%, p = .23 1.08%
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
2.4.12.3 Self-harm
We repeated the moderation analysis using dichotomised self-harm variable as the
outcome. For SP, no significant interaction effects were found between SP and self-
control (c4: B = .32, SE = .44, p = .47), or between SP and mindfulness (c5: B = -.18, SE
64
= .31, p = .56). The moderation model only explained between 4.90% (Cox & Snell R2)
or 4.97% (McFadden R2) and 7.70% (Nagelkerke R2) variance in self-harm status. Self-
harm was solely predicted by self-control (c2: B = -.77, SE = .32, p < .05), but neither
by SP (c1: B = .10, SE = .23, p = .66) nor mindfulness (c3: B = -.15, SE = .26, p = .56).
The proposed moderation model for SF revealed exactly the same pattern, such that
no significant interaction between SF and self-control (c4: B = .27, SE = .43, p = .52),
or between SF and mindfulness (c5: B = -.22, SE = .29, p = .46). The model explained
between 5.16% (Cox & Snell R2) or 5.24% (McFadden R2) and 8.11% (Nagelkerke R2)
variance in self-harm status. Here again, self-harm was only predicted by self-control
(c2: B = -.70, SE = .32, p < .05), but not by SF (c1: B = .27, SE = .23, p = .24), or
mindfulness (c3: B = -.11, SE = .26, p = .69).
Overall no support was found for Hypothesis 11b, in that only higher self-control
individuals would be less self-harmed, regardless their levels of SP, SF, or
mindfulness.
2.5 DISCUSSION
2.5.1 Relationships between mindfulness, self-control, and aggression
The primary purpose of the current study was to investigate the dynamic of
interaction between individual differences in mindfulness, self-control, and
aggression. The plausible interaction was explored by examining two different
perspectives, whether (i) self-control mediated any relationships between
mindfulness and trait aggression (ii) mindfulness accounted for incremental validity
over self-control in predicting trait aggression, and therefore might moderate any
relationships between self-control and aggression. Preliminary analysis indicated
that mindfulness was associated with reduced trait aggression, and that higher levels
of mindfulness were strongly associated with higher levels of self-control. However,
the strengths of the predictions of mindfulness and self-control on the aggression
measures were varied.
65
The proposed mediation of self-control on the link between mindfulness and general
trait aggression was supported in the current study, even after controlling for
relevant demographic factors (i.e., current educational status, sex, age,
race/ethnicity, and current mindfulness practice). This finding provides empirical
evidence that self-control may be the potential mechanism by which mindfulness
reduces aggression. Further examination of the proposed mediation model on the
four aggression subscales revealed that self-control mediated the impact of
mindfulness on physical aggression, anger, and hostility, but not on verbal
aggression. Verbal aggression appears to be more related to sex (i.e., higher levels of
verbal aggression amongst male participants).
Recently, to demonstrate the incremental validity of mindfulness over self-control
on psychological functioning, Bowlin and Baer (2012) used hierarchical regression
analyses, in which trait self-control (Tangney et al.’s SCS, 2004) was entered prior to
trait mindfulness (as measured by the FFMQ: Baer et al., 2006) at the regression
steps. With this approach, significant variance of mindfulness would be established if
the total variance of the outcome increases after the inclusion of mindfulness.
Comparable results were obtained when we rerun the analysis following the Bowlin
and Baer’s (2012) analysis (Note that in the current study, the unique contribution of
mindfulness was estimated directly with bootstrapping method, through the c’ path
[the direct effect of mindfulness on aggression]). Specifically, mindfulness showed
incremental validity over self-control in predicting trait aggression, anger, and
hostility, but not physical aggression. This finding highlights the unique role of
mindfulness in the emotional or affective component of aggression (i.e., anger) as
well as the cognitive one (i.e., hostility). The impact of mindfulness on physical
aggression, however, appears to be largely attributable to self-control mechanism.
As a comparison, one correlational study (Borders et al., 2010) found no support for
the mediating role of rumination on the link between mindfulness and physical
aggression, even though rumination mediated the link between mindfulness and
verbal aggression, anger, and hostility. Altogether, self-control may be the primary
mechanism through which mindfulness influences physical aggression.
66
Given that mindfulness accounted for significant variance in some of the aggression
measures after the inclusion of self-control, it was reasonable to expect that
mindfulness might moderate the link between self-control and aggression. No
support was found for the proposed moderation models, either on general trait
aggression or on any of the four trait aggression subscales. Higher self-controlled
individuals would be less physically aggressive and less hostile regardless of their
levels of mindfulness, whereas more mindful individuals would be less angry and
hostile despite their self-control levels. This inconclusive pattern of results implicates
the necessity to directly compare the effect of mindfulness and self-control through
experimental studies, while controlling for individual differences in aggression. To
this end, studies reported in chapters 3, 4, and 5 of the current thesis do just this,
comparing the impact of induction in mindfulness vs. training in self-control on
aggressive behaviour (Chapter 3) and the moderating effect of mindfulness on
aggressive behaviour following self-control depletion (Chapter 4 and 5).
No differences were found either in mindfulness or self-control as a function of
current mindfulness experience. It is possible that these non-significant correlations
were due to the characteristics of our sample. In the original study of the
Mindfulness Attention Awareness Scale (MAAS), Brown and Ryan (2003) recruited
meditation practitioners a Zen monastery, finding significant relationships between
the MAAS and experience with meditation. Compared to Brown and Ryan’s sample,
our meditation sample reported much lower MAAS scores (M = 3.65, SD = .91 vs. M
= 4.38, SD = .65 in our sample and the development scale sample, respectively).
Additionally, Brown and Ryan’s sample was also much older (M age = 41.08, range
22 – 62 years) than our sample (M age = 23.87, range 18 – 41), which may be a
relevant factor, as age showed positive correlation with trait mindfulness and self-
control in the current study. A recent validation study of the MAAS also suggested
for “caution in presuming experience with meditation is associated with greater
mindfulness in a general sample” (MacKillop & Anderson, 2007, p.292).
Interestingly, current mindfulness experience was inversely associated with trait
aggression and verbal aggression. One possible explanation is that those who have
undertaken mindfulness meditation might have done so in order to deal with some
67
negative emotions or similar psychological distresses, including anger and
aggression. While mindfulness practice may have helped them by reducing their
aggression levels, we have no data to assess their levels of aggression before and
after they engaged in such practice. Experimental studies could help address this
possibility by examining the differences in aggression between those who have
received mindfulness training compared to those who have not. It should also be
noted that although we have provided a clear definition of mindfulness experience in
the survey (i.e., “a specific meditational practice in directing attention without
making any judgment”), participants were only asked to rate the amount of time of
their current practice (i.e., daily, weekly, monthly, once in a year). Other crucial
indicators may have been overlooked, such as the duration of the practice history, or
the extent to which they perceived that the practice was employed into daily life
(Brown &Ryan, 2003). In addition, is also plausible that the unequal sample size
(11.2% vs. 88.8% for “experienced” and no experience groups, respectively) may
have compromised the conclusions we obtained.
Higher levels of trait aggression and anger were also reported by Asian participants
compared to White participants, whereas no ethnic differences were found in overt
aggression. This finding appears to contradict literature on the influence of cultural
values on aggression. Asian cultures are typically associated with collectivism values,
which emphasise the maintenance of social harmony and conflict avoidance,
therefore should display less tendency to aggress than societies with stronger
individualistic values (e.g., Forbes et al., 2009; Li et al., 2010; Magnis-Suseno, 1997).
Nevertheless, our data were obtained online from anonymous participants. It is
unclear whether those who classified as Asian would in reality adopt the collectivistic
or individualistic values (e.g., some might have been born and raised within Western
cultures). Hence a further cross-cultural comparison is warranted (see Chapter 6).
2.5.2 Link between aggression, mindfulness, self-control, and self-harm
Additionally, the current study aimed to examine if trait mindfulness and self-control
could predict individuals’ tendency to harm themselves in the same way as they may
68
predict harm to others. Our preliminary analysis supported that self-harm was
positively correlated to trait aggression, physical aggression, and anger, but not to
verbal aggression and hostility.
The lack of significant correlation between self-harm and hostility requires further
elaboration. Hostility towards others (as assessed with Buss & Warren’s Aggression
Questionnaire [BWAQ]: 2000) was reported as predicting the common categories of
suicidal behaviours (i.e., suicidal ideation, suicide plan, and suicide attempts)
amongst Chinese adolescents (Zhang, Roberts, Liu, Meng, Tang et al., 2012).
Participants in our study were asked to rate their self-harming behaviour in the
absence of suicidal intent only. It is plausible that the absence of this intention might
have reduced the expected association between self-harm and hostility towards
others, at least in the current sample. Measures of self-harm in the presence of
suicidal intent are noteworthy to clarify the association between self-harm and
aggression, particularly in terms of hostility.
Crucially, the current study found that self-control mediated the link between
mindfulness and self-harm in the same way as it mediated the relationship between
mindfulness and aggression. Thus information about individuals engagement in self-
harm may be useful for experimental work investigating the role of mindfulness and
self-control on aggressive behaviour.
It is particularly interesting, however, that after the inclusion of self-control,
mindfulness was no longer a significant predictor for self-harm. Likewise,
mindfulness did not moderate the link between self-harm and self-control. A similar
result was found in Slee, Spinhoven, et al.’s (2008) study using clinical population, in
which the effect of treatment on self-harming behaviour was mediated by self-
control related components (as measured by the Difficulties in Emotion Regulation
Questionnaire [DERS]: Gratz & Roemer, 2004), but not so much by the mindfulness-
related components (i.e., the awareness of and acceptance of negative emotions in
the DERS). Taken together, these finding appear to support that reduction in self-
harming behaviour occurs primarily through the mechanism of self-controlled
emotion regulation.
69
It should also be noted that in the current sample, only anxiety but not depression
predicted self-harm. Both variables have been described as the most common
factors of self-harm (e.g., Hawton et al., 2002). The non-significant correlation
between depression and self-harm may have resulted from low levels of depression
in the current sample (M depression = .63, SD = .50; M anxiety = 1.17, SD = .58). In
addition, depression levels of self-harmers were not significantly different from non
self-harmers. The presence of a floor effect could interfere with the identification of
any existing relationships. Replication with a different population is necessary to
help resolve this issue.
2.5.3 Role of individual differences in SP and SF
For a more comprehensive understanding of the predictions of aggression, the
current study explored the potential role of individual differences in sensitivity to
provocations (SP) and frustrations (SF). Results showed that both SP and SF were
positively associated with trait aggression and the four aggression subscales.
Importantly, mindfulness decreased the likelihood of anger amongst individuals who
were more sensitive to provocations and of hostility amongst those who were more
sensitive to frustrations. Conversely, self-control did not moderate any relationships
between SP and SF and overt or overall aggression. These findings strengthen the
unique role of mindfulness on anger and hostility, independent and beyond the
mechanism of self-control. The mindful awareness and acceptance of negative
emotions appear to work in particular by reducing the physiological arousal and
preparation for aggression (i.e., anger) when individuals are provoked, and by
lessening the feeling of ill will and injustice (i.e., hostility) when individuals are
trigged with frustrations.
The correlation between SP and SF and self-harm did not mirror the association
between these variables and aggression. Specifically, only SF was related to self-
harm. This suggests that SF may be a pertinent individual difference when predicting
self harm and aggression, while SP may be more specifically related to aggression.
Individuals may engage in self-harmful behaviour while dealing with situations
70
where they experienced a lack of control, but not when being provoked. Further,
moderator analysis with mindfulness and self-control revealed that self-control
predicted self-harm beyond the possible influence of SP, SF, and mindfulness. This
latter finding reiterated that self-controlled emotion regulation plays a major role in
attenuating self-harming behaviour.
2.5.4 Strengths and limitations of the current study
In the current study, all measures were self-reported and so potentially are
influenced by social desirability. To anticipate this issue, participants were told
explicitly that the study focused on how individual differences in the measures were
related to each other, not on individual profile of each participant. Moreover,
instead of counterbalancing the order of the questionnaires, we tried to minimise
participants’ barrier by presenting more sensitive measures (aggression followed by
self-harm) at the end of the survey.
A further concern is the nature of data collection. Specifically, our data were
collected online using a surveymonkey website. Despite the recommendations that
internet and traditional survey data are comparable in terms of the problems they
may impose (Gosling, Vazire, Srivastava, & John, 2004; see also Kraut et al., 2004),
one of the significant drawbacks of the internet sampling is that some participants
might want to increase their chances of winning the prize draw by entering the
survey multiple times. To anticipate this issue, at the end of the survey we asked
participants to enter some “security details” (i.e., the first letter of their hometown,
the last letter of their most favourite colour, the initial of their mothers’ maiden
name, the month they were born in, and the number of siblings they have), rather
than imposing them to provide their email addresses or to use computers with
different IP addresses.
Importantly, our main analyses relied on the bootstrapping method (Preacher & A. F.
Hayes, 2004, 2008) to test the proposed mediation and moderation models. Even
though bootstrapping method is superior in terms of power and Type I error rates
71
compared to the traditional causal approach or Sobel test (see e.g., MacKinnon et
al., 2004; Shrout & Bolger, 2002), non-experimental data can only suggest that a
proposed mediation pattern is plausible. While our hypotheses were mostly derived
from the proposal that self-control is more likely to be acquired from practicing
mindfulness mode of awareness (e.g., Bishop et al., 2004; Brown et al., 2007a;
Shapiro et al., 2006), the opposite mechanism cannot be completely ruled out.
Specifically, self-control may actually predict mindfulness—as proposed in
Masicampo and Baumeister’s (2007) theoretical work—and in turn mindfulness
predicts aggression. Fully experimental studies of causal process are of particular
necessary for mediation and moderation testing (see e.g., MacKinnon et al., 2002).
Despite the limitations, the current study provides preliminary evidence that
mindfulness might reduce aggression and self-harm through its influence on self-
control. The inclusion of measure of individual differences in SP and SF may also
provide a more comprehensive understanding about the usefulness of mindfulness
and self-control for the reductions of aggression and self-harm.
2.5.5 Conclusions and next steps
Our result suggested that both trait mindfulness and trait self-control may reduce
the general individual propensity to behave aggressively to themselves and to
others. Crucially, trait mindfulness may uniquely reduce anger and hostility
regardless of individual’s level of self-control, SP, and SF. By contrast, self-control
appears to be the primary mechanism through which mindfulness may influence
physical aggression and self-harm. In the following chapters, we will report three
experimental studies of mindfulness and self-control on aggressive behaviour in a
fully-controlled experimental design. In this way, the difficulties of interpreting cross-
sectional data with mediation analyses are minimised, and it is possible to make
some causal conclusions about the relationship between mindfulness, self control,
and aggressive behaviour. For homogeneity and ease of accessibility purposes, only
university students would be recruited as our next population.
72
CHAPTER THREE
Study 2: Effect of mindfulness induction and self-control training on aggressive
behaviour
3.1 INTRODUCTION
The results in Study 1 (Chapter 2) supported that individual differences in
mindfulness and self-control may reduce the propensity to behave aggressively.
Mindfulness also showed incremental validity over self-control in predicting trait
aggression, anger, and hostility. However, the impact of mindfulness on self-
reported physical aggression was largely attributable to self-control. Accordingly, the
current study has two major purposes. Firstly, to investigate whether experimentally
inducing state of mindfulness would reduce aggressive behaviour. Secondly, to
explore whether mindfulness induction could strenghten the potential reduction of
aggressive behaviour due to increases in self-control strength (i.e., following self-
control training). In addition, the potential roles of individual differences in self-harm
as well as in sensitivity to provocations (SP) and frustrations (SF) are also assessed.
3.1.1 Mindfulness induction and aggressive behaviour
The first aim of the current study is to examine whether reductions in aggressive
behaviour could occur by experimentally manipulating individuals’ levels of
mindfulness, using a 10-min of mindful walking exercise (Kabat-Zinn, 1994; Sujiva,
2000). Given that mindfulness is generally studied as a dispositional form and as a
result of therapeutic interventions, most aggression literature has also explored the
role of mindfulness in these contexts. As previously discussed, trait mindfulness was
shown to be related to various self-reported measures of aggression (e.g., Barnes et
al., 2007; Borders et al., 2010; Brown & Ryan, 2003; Kelly & Lambert, 2012). Our
results from Study 1 also found negative relationship between trait mindfulness and
trait aggression, physical and verbal aggression, anger, and hostility. In mental health
73
and forensic settings, several authors have recommended the potential applications
for mindfulness-based treatment of aggression and anger (see Howells, 2010;
Howells et al., 2010; Wright et al., 2009; also Singh and colleagues series of case
studies, 2003, 2007, 2011).
More recently, the benefit of mindfulness on emotion-related processes is proposed
to be observed through brief laboratory inductions, using a short guided meditation
instruction to adopt an accepting attitude toward one’s experiences (for a review,
see Keng et al., 2011). In this context, however, only one reported aggression study
(Heppner et al., 2008) actively manipulated participants’ state of mindfulness (using
a raisin eating task), providing a preliminary support for the potential effect of state
mindfulness on aggressive behaviour (a description of this study is provided in
section 1.2.2.4, Chapter 1). As in Heppner et al.’s study, our current study employs
the same measure of aggressive behaviour called the Taylor Competitive Reaction
Time (TCRT: Taylor, 1967) task, specifically in its adapted form.
In typical work using the TCRT task, participants play a computer-based competitive
reaction-time task against a bogus partner, where the winner of each trial is given
opportunity to deliver a noise blast/electric shock to the loser. In the adapted
version of the TCRT task, a non-aggressive option is provided for the participants
(e.g., De Wall, Buckner, Lambert, Cohen, & Fincham, 2010; Konrath, Bushman &
Campbell, 2006; Muller, Bushman, Subra, & Ceaux, 2012; for critics on the
importance of providing non-aggressive response options, see Ritter & Eslea, 2005;
Tedeschi & Quigley, 1996). Following past research (Anderson, Buckley, & Carnagey,
2008; Lawrence & Hutchinson, 2013a, 2013b), we use participant’s intensity of the
noise blasts to the bogus opponent as a main measure of direct aggression.
Bettencourt, Talley, Benjamin, and Valentine’s (2006) meta-analytic review
concludes that the influence of personality variables on aggressive behaviour may
differ under provoking and nonprovoking conditions. This indicates the necessity for
the current study to distinguish the proposed effect of mindfulness induction on
provoked and unprovoked aggression, and account for the potential role of
individual differences (trait mindfulness, self-control, aggression, self-harm, SP, and
74
SF). While higher blast intensities are commonly associated with increased levels of
provocation, interventions aimed at reducing aggression in the TCRT task tend to
lose their effectiveness under high provocations (see Lawrence & Hutchinson, 2013a,
2013b). Thus instead of using a social rejection task in Hepper et al.’s (2008) study to
provoke aggression, we vary the levels of provocation in the TCRT task, such that
participants are exposed to no provocation, low provocation, and high provocation
trials. By doing so, it is possible to see whether any impact of mindfulness induction
would persist despite opponents’ increases in levels of provocation.
As concluded by Giancola and Parrot (2008), it is more likely for people to engage in
implicit aggression (represented by shock duration in the TCRT task) than explicit
aggressions (shock intensity) when their aggressive impulses are inhibited by other
factors, such as cultural values and sex role norms. In the current study, we include
measure of the maximum latency (the delay duration before the maximum blast was
delivered the opponent in the TCRT task; see Lawrence & Hutchinson, 2013a,
2013b), which is arguably similar to shock duration in its more subtle form of
aggression. In this way, the potential influence of mindfulness induction on different
types of aggression can be explored.
We also assess aggressive behaviour in terms of indirect aggression, specifically
through participants’ anonymous reputation damage of the opponent. A negative
judgment would reflect an immediate intent to cause harm to the target, which
corresponds to Bushman and Anderson’s (2001) widely accepted definition of
aggression. A similar method has been used as a sole measure of aggressive
behaviour in many previous studies (e.g., De Wall et al., 2007; Stucke & Baumeister,
2006; Twenge, Baumeister, Tice, & Stucke, 2001). While higher levels of direct
aggression in the TCRT task have been demonstrated in males than in females
(Giancola & Parrott, 2008), sex differences in aggression typically narrows when
indirect aggression is measured (see Card, Stucky, Sawalani, & Little, 2008).
In addition, the current study examines the proposed effect of mindfulness induction
on a behavioural measure of self-control performance as an additional dependent
variable beyond aggressive behaviour. Specifically, we use performance in a
75
handgrip task that has been identified as one of the frequently used dependent task
in Hagger’s et al. (2010) meta-analysis. We expect that the immediate effect of
mindfulness induction on self-control performance would be sustained after
participants have engaged in a provocation procedure.
3.1.2 Self-control training and aggressive behaviour
In a validity study of the Taylor paradigm, Giancola and Parrott (2008) demonstrated
that direct aggression during the TCRT task were mostly associated with trait
physical aggression (assessed using Buss & Perry’s AQ, 1992). Results in Study 1
revealed the influence of trait mindfulness on self-reported aggression, physical
aggression, anger, and hostility was mediated by trait self-control. After the inclusion
of trait self-control, however, the predicting role of trait mindfulness on physical
aggression was no longer significant). It is particularly crucial, therefore, to
determine whether the effect of state induced mindfulness on aggressive behaviour
could indeed be separated from the effect of self-control. The next aim of the
current study is to test whether (i) increasing self-control strength reduces
aggressive behaviour in the same way it could be reduced by mindfulness induction,
and (ii) the potential reduction of aggressive behaviour due to increases in self-
control strength could be strenghtened by mindfulness induction.
As introduced in Chapter 1 (section 1.2.3.2), the majority of self-control experiments
appear to fall under the strength model of self-control advocated by Baumeister and
colleagues (see Hagger et al.’s meta-analysis, 2010). Here, different acts of self-
control are proposed to draw from a common, general “resource” (Baumeister,
Vohs, et al., 2007). Performing self-control acts may temporarily deplete this
resource (this phenomenon will be investigated in the subsequent studies of this
thesis). Conversely, regular practice of self-control over extended periods can
improve the general capacity for self-control.
Outside the area of aggression, the effects of various self-control training (e.g.,
improving posture, tracking eating habit, using nondominant hand in a mundane
76
task, avoiding colloquialisms/cursing) is usually demonstrated in the form of
increases in stamina or a greater resistance to ego-depletion (e.g., Muraven et al.,
1999; Gailliot et al., 2007; Oaten & Cheng, 2006a, 2006b, 2007). That is, individuals’
performance on a self-control task following ego-depleting task is measured twice, at
baseline level and after a period of training in a typically unrelated self-control task.
The effect of self-control training is assessed by comparing the degree of impairment
in self-control performance following depletion between self-control trained and
non trained individuals at both time points (see Baumeister et al.’s, 2006, review).
More recently, Muraven (2010) found that individuals who practiced self-control
(i.e., avoiding sweets or squeezing a handgrip) for two weeks showed significant
improvement in self-control performance (i.e., a stop signal task) compared to those
who practiced tasks that did not require self-control (i.e., solving math problems or
keeping a diary of any self-control acts they engaged in). Thus consistent with the
ways that muscular strength can be improved, increases in self-control strength
could also occur in terms of self-control power. The current study employs self-
control training of regulating posture for a period of two weeks (Muraven et al.,
1999), and examines the effect of this training on self-control power in a handgrip
task. We expect that self-control training participants would show a better handgrip
power compared to non training participants.
Importantly, the effect of self-control training is also tested on aggressive behaviour.
Corresponding with the view that self-control serves to restrain aggressive impulses
(see Denson et al., 2012; DeWall et al., 2011), training in self-control has been shown
to decrease aggression, particularly following provocation. Denson et al. (2011)
found that a two-week training in physical regulation (using nondominant hand in
mundane tasks) reduced direct aggression in the TCRT task (i.e., the mean score of
blast intensity and duration) in response to insulting feedback from a bogus
opponent, but only amongst individuals who scored higher in trait physical
aggression (for details see section 1.2.3.3, Chapter 1). The same physical training,
and a verbal regulation training of avoiding colloquialisms were used in Finkel et al.’s
(2009) study involving a hypothetical scenario of provocations from an intimate
partner. Specifically, all participants were asked to perform an ego-depleting task
77
(i.e., attention control task) before they completed a self-reported measure of
violent inclinations towards a provoking partner—at baseline level and after
performing self-control training. As predicted, self-control trained participants
showed less violent inclination over time whereas no reductions were found in non
self-control trained participants.
As mentioned in the earlier section, the current study adds a no-provocation
condition, thus the impact of self-control training can be measured on both
provoked and unprovoked aggression. In addition, self-control training is expected to
influence other dependent measures, i.e., the maximum blast latency, rating of
indirect aggression, and self-control performance in the same way these measures
are proposed to be influenced by mindfulness induction. Moreover, we expect that
mindfulness induction could strengthen the effect of self-control training on
aggressive behaviour and self-control performance.
Since Study 1 found that trait mindfulness and self-control reduced individuals’
tendency to harm themselves in a similar way as they predicted measures of trait
aggression (section 2.4.8 and 2.4.9 in Chapter 2), the impact of mindfulness
induction and self-control training could also be influenced by self-harm status. The
impact of these manipulations may also be affected by individual differences in
sensitivity to provocations (SP) and frustrations (SF). In the adapted TCRT task,
individual differences in SP has been shown to be positively related to blast intensity
following provoked aggression and to the maximum blast latency, but not to indirect
aggression (Lawrence & Hutchinson, 20013b). Thus measures of individual
differences in mindfulness, self-control, aggression, SP, SF, and self-harm are also
assessed in the current study, along with relevant demographic factors, i.e., sex,
race/ethnicity, and participants’ current practice of mindfulness.
3.2 HYPOTHESES
The primary hypotheses of this study are concerned with the role of mindfulness
induction and self-control training on aggressive behaviour, and predict that:
78
1. Compared to those not receiving mindfulness induction, participants receiving
mindfulness induction will:
a. Deliver lower levels of blast intensity to the opponent, particularly on low
provocation trials. No strong prediction is made regarding the levels of
blasts delivered under no provocation or high provocation trials.
b. Wait longer before delivering the maximum blasts to the opponent.
c. Deliver lower ratings of anonymous reputation damage to the opponent.
2. Those who receive self-control training will be less aggressive than those without
self-control training, thus mirroring the associations with mindfulness induction
(Hypothesis 2a – 2c).
3. Mindfulness induction will strengthen the effect of self-control training on
aggressive behaviour. Specifically, compared to self-control training participants
with no mindfulness induction, self-control training participants who also receive
mindfulness induction will:
a. Deliver lower levels of blast intensity to the opponent, particularly on low
provocation trials. No strong prediction is made under no provocation or
high provocation trials.
b. Wait longer before delivering the maximum blasts to the opponent.
c. Deliver lower ratings of anonymous reputation damage to the opponent.
4. The moderation of mindfulness on the link between self-control training and
aggressive behaviour will persist after accounting for the potential influence of
individual differences in mindfulness, self-control, aggression (i.e., trait
aggression, sensitivity to provocations [SP] and frustrations [SF]), sex, self-harm,
ethnicity/culture, and current mindfulness practice.
In addition, we predict that:
5. Mindfulness induction will strengthen the effect of self-control training on self-
control performance, as measured by handgrip power after accounting for trait
and demographic covariates.
79
As a whole, we hypothesise that self-control training and mindfulness induction will
predict lower levels of aggression and a better self-control (handgrip) performance,
after accounting for covariates (personality variables). Moreover, mindfulness
induction will strengthen the effect of self-control training on aggression.
3.3 METHODS
3.3.1 Participants
University of Nottingham students took part in two times points of the current study.
A total of 147 (72 females) students entered Time 1 of the study. Based on sex,
participants were randomly assigned to one of four groups (see randomisation
details in the next section). Data were removed from further analyses due to: not
attending the Time 2 session (n = 19 from the self-control training condition and n =
18 from the no training condition), not providing any compliance scores to the self-
control training (4 participants), and expressing spontaneous suspicions regarding
the TCRT task (7 participants). Our final sample consisted of 99 participants (51
females). Participants’ mean age was 20.02 (SD = 2.66). The ethnic distribution was
61.6% White, 29.3% Asian, 7.1% participants rated themselves as “Others”, and 2.0%
did not report their ethnical background. Concerning current mindfulness practice,
91.9% participants reported never having encountered mindfulness, 3.0% practised
them on a monthly basis, 4.0% on weekly basis, and 1% (1 participant) practised
them on daily basis. The experiment lasted from October until November 2011.
3.3.2 Design
To these ends, we conducted a 2 (self-control training vs. no self-control training) x 2
(mindfulness induction vs. relaxation) experimental design (See Figure 3.1 for the
flow of participants). Stratified sampling technique was employed to assess the
effect of sex, as sex differences in aggression are prevalent in studies using the TCRT
tasks (e.g., Anderson et al., 2008; Giancola & Parrott, 2008).
80
Figure 3.1. Flow of participants through Study 2.Note. M = Males; F = Females
3.3.3 Trait and state measures
At Time 1 of the current study, participants were asked to complete the self-
reported measures in Study 1 (Appendix 2.1, the full description of the trait
measures can be found in section 2.3.3, Chapter 2). Specifically, we assess individual
differences in mindfulness (Mindfulness Attention Awareness Scale [MAAS]: Brown
& Ryan, 2003), self-control (Self-Control Scale [SCS]: Tangney et al., 2004), aggression
(Aggression Questionnaire [AQ]: Buss & Perry, 1992), sensitivity to frustrations and
provocations (Situational Triggers of Aggressive Responses [STAR scale]: Lawrence,
2006), and self-harm (Deliberate Self-Harm Inventory [DSHI]: Gratz, 2001).
The measure of anxiety and depression (Hospital Anxiety and Depression [HADS]:
Zigmond & Snaith, 1983]) from Study 1 was replaced with a mood measure (Positive
Affect, Negative Affect Schedule [PANAS]: Watson, Clark & Tellegen, 1988). The
PANAS consists of two 10-item mood scales measuring positive affect (10 items, e.g.,
81
“interested”, “distress”) and negative affect (10 items, e.g., “irritable”, “ashamed”).
Participants were asked to rate on a 5-point scale (1 = very slightly or not at all and 5
extremely) the extent to which they experienced each mood state (Appendix 3.1). It
is important to employ this measure following the TCRT task, since negative affect
due to unpleasant experiences often trigger aggressive behaviours (Anderson &
Bushman, 2002; Berkowitz, 1993, 2008).
The efficacy of the mindfulness intervention was assessed using Toronto
Mindfulness Scale (TMS: Lau et al., 2006; see also Appendix 3.1). The TMS is based
on Bishop et al.’s (2004) two-component model of mindfulness. It consists of 13
items assessing curiosity (6 items, e.g., “I was curious to see what my mind was up to
from moment to moment”) and decentering (7 items, e.g., ‘‘I was more concerned
with being open to my experiences than controlling or changing them.’’), on a 5-
point Likert-scale (0 = not at all and 4 = very much). Curiosity refers to the awareness
of present experience with a quality of curiosity, while decentering refers to the
awareness of one’s experience with some distance and disidentification rather than
being carried away by one’s thoughts and feelings (Lau et al., 2006). Currently, the
TMS is the only scale measuring mindfulness as a state (Bergomi, Tschacher, &
Kupper, 2012; Sauer et al., 2013; see Feldman, Greeson & Senville, 2010; Ortner,
Kilner, & Zelazo, 2007; Thompson & Waltz, 2007 for laboratory work using the TMS).
For the self-control trained group, we asked participants to rate how frequently they
comply with the instruction of maintaining good posture (e.g., sit up straight, walk
erectly, etc.) at all time, on a 6 point Likert-scale (1 = almost never and 6 = almost
always). The mean score of compliance rating in two weeks period of training was
used to determine participants’ successfulness in following the instruction.
3.3.4 Experimental manipulations
3.3.4.1 Self-control training
Those allocated to self-control training were instructed to perform a two-week task
of improving posture. Specifically, participants were asked to monitor and correct
82
their postures (e.g., sit up straight, walk erectly, etc.) at all times. They were told that
in order to gain the benefit of an improved posture, it is crucial for them to exert as
much control as possible. Daily remainders for this group were sent via email, along
with a single question about how compliant they were with the training (Appendix
3.2). They were given alternatives to reply to the emails with their compliance score
on a daily basis, or to hand in the print out version on the day of the experiment (at
Time 2 of the current study). Those in the no-training control group were not
required to perform any additional activities from usual during this interim period. A
similar training has been reported to increase handgrip performance following
thought-suppression compared to no training (Muraven et al., 1999).
3.3.4.2 Mindfulness induction
For the mindfulness induction, we used a 10-min of mindful walking exercise (Kabat-
Zinn, 1994; Sujiva, 2000; see Appendix 3.3). Participants were asked to intentionally
direct their attention towards witnessing the full experience or the sensations of
walking (the movements, shifts of weight and balance, and sensations in the feet and
legs), as opposed to the physical feet per se. Variants of these instructions were
repeated every 60 s for 10 minutes. This exercise is considered as both an example
of mindfulness exercises in daily life and a formal meditation practice to cultivate
mindfulness (Siegel, Germer, & Olendzki, 2009). Its benefits include (i) people with
difficulty to stay still for a long time may find walking to be easier to help them
develop attention and awareness, (ii) the bodily experience of walking provides
more clear and vivid subject for meditation than meditation while sitting and lying,
(iii) body awareness is in fact the first foundation of mindfulness (see Bien &
Didonna, 2009). Even though empirical evidence of the efficacy of this exercise per
se seems to be lacking, it has been formally integrated in the Mindfulness-Based
Stress Reduction (Centre for Mindfulness, 1995) and Mindfulness-Based Cognitive
Therapy (Oxford Mindfulness Centre, 2002).
For the control condition, we used the Progressive Muscle Relaxation (PMR:
Jacobson, 1938; see Appendix 3.4) exercise. Specifically, participants were guided by
83
a 10-min audio instruction to tense each muscle group vigorously but without
straining, and then suddenly release the tension and feel the muscle relax. This
exercise was given to rule out the possibility that relaxation alone could account for
the effects of mindfulness induction—since mindfulness training is modelled after
meditation exercise, which often leads to not only sustained attention on the
present moment, but also relaxation (Baer, 2003).
3.3.4.3 Direct aggressive behaviour
We used the adapted Taylor Competitive Reaction Time (TCRT: Taylor, 1967) to
provoke participants’ aggressive behaviour, using E-Prime software. The TCRT task
has been employed to measure the effect of self-control training (Denson et al.,
2011) and mindfulness induction (Heppner et al., 2008) on aggressive behaviour.
Participants were informed that they would be playing a computer-based
competition task against an opponent located in a nearby room —In reality there
was no opponent and the outcome of the task was pre-programmed. The task was
presented as a series of reaction-time trials, in which participants have to respond
(i.e., hit the spacebar on the computer keyboard) as fast as they can when a white
circle stimuli appeared on the screen. Each time they won a trial, they were asked to
deliver a noise blast to the opponent, by selecting a level ranging from 0 (no blast,
non-aggressive option) through to 8 (maximum blast). When they lost a trial, they
received a noise blast (lasted for 500 ms) through surround earphones at a level
chosen by the opponent. The maximum blast level delivered to the participants was
set according to their own threshold. Specifically, prior to the TCRT task, participants
indicated the maximum intensity they were prepared to accept (up to a maximum of
90 db). This maximum level was then reduced in units of 5 db from 8 to 1.
In the TCRT task, participants always won the first trial to assess baseline blast
intensity that they delivered without any provocation from the opponent. On
subsequent losing trials, they received blasts in two blocks, i.e., low intensity (40
trials, in which participants lost 50% of time at random and received blast levels 1 to
84
4) and higher intensity (40 trials, in which participants lost 50% of time at random
and received blast levels 5 to 8). Hence, participants experienced no provocation,
low provocation, and high provocation trials.
The intensity of blasts delivered to the opponent under each block of trials (no
provocation, mean low provocation, and mean high provocation) was used as a
measure of direct aggression (see e.g., Giancola & Parrott, 2008; for further evidence
of the validity of this paradigm). Due to restrictions by the Local Ethics Committee,
blast duration typically measured in the TCRT studies was not assessed in the current
study. Instead, we used the maximum blast latency (i.e., the number of trials
participants waited before delivering the maximum blast; see Lawrence &
Hutchinson, 2013a, 2013b) as an additional measure of direct aggression.
3.3.4.5 Indirect aggression
To avoid suspicions that the participants were playing against a computer, and
moreover to provide alternative ways of aggression measures, at the end of the
TCRT task participants were asked to make an evaluation of the opponent.
Participants made their ratings on how aggressive, skilful, competitive, and fair they
opponents were, and on their willingness to play against the same opponent again in
future studies on a 7-point scale (1 = completely disagree and 7 = completely agree).
Prior to the evaluation, participants were told that their judgment would help the
experimenter decide whether to use the opponent in future studies with payment or
not, and that their evaluation would be kept confidential. We called this evaluation
“indirect aggression” (see Lawrence & Hutchinson, 2013a 2013b).
3.3.4.6 Self-control performance (Handgrip task)
Self-control performance was measured using a handgrip task. This is a well-
established measure of self-control which compromises resisting fatigue and
overriding the urge to quit (see Hagger et al., 2010). We used a commercially
85
available digital hand grip dynamometer designed for measuring the flexion of the
wrist and an accurate reading of grip strength using the dominant hand. Since we
were interested to measure not the physical power but the self-control power, we
asked participants to continuously squeeze the handgrip for as long as they could
and timed how long the participant held the handles squeezed.
3.3.5 Power calculation
Due to a lack of previous similar study, a precise effect size for this current study
could not be predicted. A total 128 participants is required to detect a medium
effect (d = .25) for four groups and 1 covariate (sex), and given the power of .80 and
an alpha level of .05 as calculated by G*Power 3 (2013).
3.3.6 Procedure
Potential participants from University of Nottingham (UoN) were recruited for the
current study via posters/leaflets on campus and departmental mailing list.
Participants were told the study was investigating the effects of personality traits
and experimental treatments on the way people perform in a competitive reaction-
time task (Appendix 3.5). Crucially, they were asked to take part in two times point,
and were informed that following the Time 1 of the study, they may or may not be
randomly allocated to perform a two-week exercise of maintaining a good posture
afterwards. At Time 1, participants were given a chance to win £25 prize draw. At
Time 2, they were given inconvenience allowance of £5 and a chance to win £25 for
the fastest participants’ reaction time across all participants.
Time 1. Interested participants were replied with one of the alternative four website
links (each link represented a different group of experiment in the current study).
Similar to Study 1, interested participants were first shown an electronic information
sheet indicating the purpose of the study (Appendix 3.6). After giving informed
consent, they were asked to provide demographic information (i.e., sex, age,
86
ethnicity, and current mindfulness practice) and to complete the online version of
the trait measures (see section 3.3.3 and Appendix 2.2). Upon completion of the
survey, participants were asked to enter an email address to win a £25 prize draw.
They were also told that this email address would be used for further invitation to
Time 2 of the study; however, their lack of participation in Time 2 did not affect their
entry into the incentive prize draw from Time 1.
Two-week interim. At the end of the online survey, those allocated to self-control
training were instructed to perform a two-week task of improving posture (Appendix
3.7). Specifically, participants were asked to monitor and correct their postures (e.g.,
sit up straight, walk erectly, etc.) at all times. Daily remainders were sent via email,
along with a single question about how compliance they were with the training (see
Appendix 3.2). Some participants replied to the emails with their compliance scores
on a daily basis, whereas others handed in the print out version at Time 2. No
additional activities were imposed for those in the no self-control training.
Time 2. Two weeks after Time 1, participants came into the lab in the School of
Psychology individually. Information about this part of the study was presented
(Appendix 3.8), along with consent form (Appendix 3.9). Half of those in the self-
control training group and half in the no training condition were asked to watch a 3
min video demonstration of mindfulness walking exercise, whereas the rest of the
participants watched the relaxation exercise video. Then experimenter left the room,
while participants listened to and practised the appropriate audio guided instruction
for 10 minutes.
The experimenter entered the cubicle door at the end of 10 minutes, and asked
them to complete the Toronto Mindfulness Scale (TMS, Lau et al., 2006). Then the
experimenter left the room for a while ostensibly to prepare their TCRT task
opponent. When the experimenter returned to the room, they told the participants
that their opponent was now ready to play against them. Then the participants were
left alone to perform the TCRT task. When the participant opened the cubicle door
to indicate the end of the TCRT, they were asked to complete the PANAS (Watson et
87
al., 1988) and to perform a handgrip task. At the end of the study, participants were
debriefed and given help and support information (Appendix 3.10).
3.3.7 Data analysis
Associations between personality variables, and between personality variables and
the experimental outcomes (i.e., aggressive behaviour and self-control performance)
were examined using zero-order correlations, independent t-tests and one-way
ANOVAs as appropriate. The efficacy of the experimental conditions was assessed
using independent t-tests or one-way ANOVAs. These statistical analyses were
performed using IBM SPSS statistics 20.0 (2011).
3.3.7.1 Effect of mindfulness induction and self-control training on aggressive
behaviour
A bootstrapping method using model “number 1” in the A. F. Hayes’s (2012a)
PROCESS macro for simple moderation (see section 2.3.6.1, Chapter 2) was
employed for the primary hypotheses. In this proposed model, self-control training
was used as predictor, mindfulness induction as moderator, and aggressive
behaviour as outcome. Aggressive behaviour was measured in terms of (i) blast
intensity on each provocation level (i.e., no provocation, low provocation, high
provocation), (ii) maximum blast latency (i.e., the number of trials participants
waited before delivering the maximum blast to the opponent), and (iii) rating
indirect aggression (i.e., anonymous reputation damage of the opponent’s
aggressiveness, skilfulness, competitiveness, fairness, participants willingness to play
against the same opponent again in).
The proposed effect of mindfulness induction on the link between self-control
training and blast intensity (Hypothesis 3a) and rating indirect aggression
(Hypothesis 3c) would be supported if the interaction coefficient between self-
control training and mindfulness induction is statistically different from zero. For
indirect aggression, we tested if the effect would survive Bonferroni corrections set
at p < .05/5 items (or p < .01).
88
Hypothesis 1a, 1c and 2a, 2c would be confirmed if the proposed moderation model
provides a significant main effect of mindfulness induction and self-control training
on blast intensity and ratings of the opponent.
For the maximum blast latency, we conducted a censored survival analysis using Cox
regressions to assess the extent to which participants would be more likely to
administer the maximum blast to the opponent as a function of experimental
conditions. We expected that maximum blast latency would be predicted by
mindfulness induction (Hypothesis 1b), self-control training (Hypothesis 2b), and the
interaction between self-control training and mindfulness induction (Hypothesis 3b).
For Hypothesis 4, the moderation models above would be repeated while accounting
for the potential influence of the hypothesised covariates, based on the significant
zero-order correlations between personality variables and aggressive behaviour.
3.3.7.2 Effect of mindfulness induction and self-control training on self-control
performance
The proposed role of mindfulness induction on the link between self-control training
and self-control performance (handgrip power) was also examined using
bootstrapping method for simple moderation. We expected that the interaction
coefficient between self-control training and mindfulness induction would be
statistically different from zero after accounting for the potential influence of the
hypothesised covariates, based on the prior significant zero-order correlations
between covariates and self-control performance (Hypothesis 5).
3.4 RESULTS
3.4.1 Baseline levels
The four experimental groups (i.e., self-control training plus mindfulness, self-control
training no mindfulness, no self-control training mindfulness, no self-control training
89
no mindfulness) were not significantly different from each other in means of the trait
measures (p = .37 to .99), indicating groups equivalency at baseline levels.
3.4.2 Preliminary analysis
The zero-order correlations amongst trait measures and descriptive statistics are
presented in Table 3.1 (N = 99). All measures showed good internal reliabilities ( =
.79 to .80). Crucially, replicating the findings from Study 1 (section 2.4.4, Chapter 2),
trait mindfulness and trait self-control were positively correlated to each other and
negatively correlated to the aggression measures.
Table 3.1.Zero order correlations and psychometric properties of measures
Measures 1 2 3 4 5 6 7 8 9
MAAS (1) 1.00
SCS (2) .52***
1.00
Total AQ (3) -.42***
-.39***
1.00
AQ Physical (4) -.29**
-.24*
.82***
1.00
AQ Verbal (5) -.22*
-.21*
.79***
.58***
1.00
AQ Anger (6) -.37***
-.42***
.86***
.58***
.59***
1.00
AQ Hostility (7) -.47***
-.43***
.79***
.55***
.40***
.59***
1.00
STAR Provocations (8) -.36***
-.28**
.50***
34***
.25*
46***
55***
1.00
STAR Frustrations (9) -.35***
-.32**
.53***
.39***
.25*
.49***
.59***
.73***
1.00
M 3.69 3.13 2.53 2.14 2.83 2.41 2.74 2.87 2.54
SD .83 .59 .63 .71 .76 .83 .79 .82 .82
Cronbach .88 .80 .90 79 .82 .87 .90 .77 .73
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression
Questionnaire; STAR = Situational Triggers of Aggressive Responses*p < .05;
**p < .01;
***p < .001
Additionally, females (n = 51) and males (n = 48) did not differ in any trait measures
(p = .07 to .73). Compared to non self-harmers (n = 85), self-harmers (n = 14) scored
lower on trait mindfulness (t(97) = -1.95, p <.05; M self-harmers = 3.29, SD =.79 vs. M
non self-harmers = 3.75, SD =.82), but higher on trait aggression (t(97) = 3.23, p <
.01; M self-harmers = 3.01, SD = .71 vs. M non self-harmers = 2.45, SD = .58), physical
aggression (t(97) = 3.41, p < .001; M self-harmers = 2.71, SD = .83 vs. M non self-
harmers = 2.05, SD = .65), hostility (t(97) = 3.38, p < .001; M self-harmers = 3.37, SD =
90
.83 vs. M non self-harmers = 2.64, SD = .73), sensitivity to provocations (SP: t(97) =
2.10, p < .05; M self-harmers = 3.29, SD = .95 vs. M non self-harmers = 2.80, SD =
.78), and sensitivity to frustrations (SF: t(97) = 2.59, p < .01; M self-harmers = 3.05,
SD = .89 vs. M non self-harmers = 2.45, SD = .78).
Due to a small number of participants who currently engaged in mindfulness practice
(n = 8 out of 99), this variable was dichotomised, resulting in a positive association
between current mindfulness practice and trait self-control (t(97) = 2.02, p < .05; M
with mindfulness practice = 3.52, SD = .36 vs. M with no mindfulness practice = 3.09,
SD = .60), and an approaching significant relation with trait physical aggression (t(97)
= 1.97, p = .052; M with mindfulness practice = 2.61, SD = .66 vs. M with no practice
= 2.10, SD = .70). Ethnicity was linked to trait mindfulness (F(2,94) = 4.15, p < .05; M
White = 3.53, SD = .76 vs. M Asian = 3.96, SD = .90 vs. M “Others” = 4.18, SD = .69)
and to self-harm status (2(2) = 7.38, p < .05; n White = 12 [85.7% of the total that
self-harmed] vs. n Asian = 0 vs. n “Others” = 2 [14.3% of the total]).
When we ran partial correlations controlling for self-harm status, the correlation
between trait mindfulness and verbal aggression fell just short of significance (r = -
.18, p = .06), and so did the relationship between trait self-control and verbal
aggression (r = -.20, p = .051). None of the associations were significantly altered
when we controlled participants’ sex (r = -.21 to .88), ethnicity (r = -.22 to .86), or
current mindfulness practice (r = -.22 to .87).
3.4.3 Zero-order correlations between personality variables and experimental
outcomes
Table 3.2 depicts the associations between personality variables and experimental
outcomes (i) direct aggression: blast intensity on each provocation level and
maximum blast latency, (ii) five ratings of indirect aggression, and (iii) self-control
performance/handgrip power. Significantly associated personality variables would
be controlled in further analyses.
91
Table 3.2.
Zero order correlations of personality measures, aggressive behaviour, and self-control performance
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression Questionnaire; STAR = Situational Triggers of Aggressive Responses; TMS =Toronto Mindfulness Scale, PANAS 1 = Positive Affect Negative Affect following depletion; PANAS 2 = Positive Affect Negative Affect after TCRT taskAll correlations were based on N = 99, except for maximum blast latency. The correlations for maximum blast latency was made only for participants who delivered amaximum blast (N = 83).*p < .05;
**p < .01;
***p < .001
92
3.4.3.1 Associations with direct aggression
Individual differences in SP were positively associated with provoked direct
aggression, but not to unprovoked aggression. Direct aggression under all levels of
provocation was related, positively, to positive affect (PA). Direct aggression did not
differ as a function of participants’ sex (p = .17 to .37; see Table 3.3), self-harm status
(p = .28 to .92), ethnicity (p = .18 to .25), or mindfulness practice (p = .49 to .89).
Higher trait physical aggression was associated with earlier deliverance of the
maximum blasts to the opponent. The likelihood of delivering the maximum blast
sooner was approaching significant for self-harmers (n = 11) compared to non self-
harmers (n = 72; t(81) = 1.88, p = .06; M self-harmers = 7.55, SD = 9.62 vs. M non
self-harmers = 13.11, SD = 9.07). No differences on maximum blast latency were
found between males (n = 39) and females (n = 44; t(81) = -1.25, p = .22), between
those who currently practice mindfulness (n = 5) and those without these practices
(n = 78; t(81) = 1.65, p = .10) or between ethnicities (F(2,79) = .67, p = .52).
3.4.3.2 Associations with indirect aggression
The rating of opponent skill was negatively related to trait mindfulness and positively
related to curiosity. The rating of opponent competitiveness and fairness was
positively related to PA. None of the indirect aggression ratings differed as a function
of participants’ sex (p = .20 to .95), self-harm status (p = .36 to .91), ethnicity (p = .10
to .87), or current mindfulness practice (p = .10 to .76).
3.4.3.3 Associations with self-control performance
Self-control performance (i.e., handgrip power) was negatively related to SP. A
better handgrip performance was shown by males compared to females (t(83.17) =
3.61, p < .001; M male = 111.65, SD = 87.87 vs. M female = 56.53, SD = 60.64), and by
those who classified themselves as belonging to “Others” ethnicity background (M
“Others” = 162.14, SD = 145.66) compared to White (M = 84.72, SD = 74.67) or Asian
93
(M = 64.69, SD = 60.47; F(2.94) = 4.46, p < .05). Handgrip power did not differ as a
function of self-harm status (p = .58), or current mindfulness practice (p = .24).
3.4.4 Manipulation Checks
To determine whether participants in the self-control training (n = 49) were
modifying their posture during the 2-week interim, we examined their responses in
the journals. The majority of participants reported compliance with the self-control
training instructions to an average level (M = 3.47 out of 6, SD = .92), suggesting that
they were relatively successful in following the instructions.
Participants in mindfulness walking exercise (n = 49) did not report higher curiosity
(t(97) =.87, p = .38; M mindfulness induction = 2.20, SD = .89 vs. M no induction =
2.04, SD = .94), or decentering (t(97) = 1.12, p = .31; M mindfulness induction = 2.07,
SD = .63 vs. M no induction = 1.94, SD = .64) than relaxation participants (n = 50).
Thus differential effect between mindfulness induction and relaxation should be
carefully interpreted as both conditions induced a comparable level of state
mindfulness. Internal consistency were good for curiosity ( = .86) and fair for
decentering ( = .67). Both subscales were positively correlated (r = .37, p < .001).
Following the TCRT task, no mood differences were found between the four
between subjects groups on PA (F(3,95) = 1.27, p = .29), or NA (F(3,95) = .29, p = .83).
Both mood subscales showed good internal consistency (= .84 and .77 for PA and
NA, respectively), and were positively correlated (r = .22, p < .05).
3.4.5 Effect of mindfulness induction and self-control training on direct
aggression
3.4.5.1 Levels of blast intensity
Since all participants experienced no provocation, low provocation, and high
provocation trials, we used a one-way repeated-measure ANOVAs to test if each trial
94
provoked different responses. Blast intensity was significantly affected by level of
provocation (F(1.38, 135.33) = 22.90, p < .0001; Figure 3.2). Specifically, more
intense blasts were shown on high provocation (M = 4.45, SD = 1.99) compared to
low provocation (M = 3.54, SD = 1.81; p < .0001), and borderline differences were
shown on low provocation compared to no provocation (p = .055; M = 3.13, SD =
2.71).
Figure 3.2. Effect of level of provocations on blast intensity.Error bars indicate standard error of mean. Scale goes up to 8 as a maximum.
When we included sex a 2 (sex) by 3 (provocation level) mixed analysis ANOVA, blast
intensity was still affected by provocation levels (F(1.40, 135.95) = 23.38, p < .0001),
and by provocation level x sex (F(1.40, 135.95) = 5.38, p < .05). Specifically, sex
differences occurred under no provocation compared to low provocation (p < .01),
but not under low provocation compared to high provocation (p = .85). In other
words, females (compared with males) delivered more intense blasts as the level of
provocations increased. The main effect of sex was not significant (Table 3.3).
Table 3.3.Sex differences on blast intensity on each provocation level
Level of provocationsMales (n = 48) Females (n = 51)
tM (SD) M (SD)
No provocation 3.52 (2.92) 2.76 (2. 46) t(97) = 1.40, p = .17
Low provocation 3.37 (1.99) 3.70 (1.63) t(90.87) = -.90, p = .37
High provocation 4.26 (2.10) 4.64 (1.88) t(97) = -.94, p = .35*p < .05;
**p < .01;
***p < .001
95
For the main hypothesis, the mindfulness induction x self-control training model on
blast intensity was tested across three provocation levels, using a bootstrapping
method with 95% bias-corrected confidence intervals (based on 5,000 bootstrap
resamples, N = 110; see Table 3.4 and Figure 3.3). All predictors were mean-
centered. Contrary to prediction (Hypothesis 1a), mindfulness induction did not
influence blast intensity. Supporting Hypothesis 2a, self-control training influenced
provoked, but not unprovoked aggression. Specifically, self-control trained
participants showed lower levels of blast intensity under low and high provocation
trials than non self-control trained participants. No support was found for
Hypothesis 3a, in that there was no significant effect of mindfulness induction on the
link between self-control and blast intensity under all levels of provocation. Self-
control trained participants who received mindfulness induction delivered a
comparable level of blast intensity with self-control trained ones with no
mindfulness induction under all levels of provocation.
Figure 3.3. Mindfulness x self-control training effect on blast intensity. Error bars indicatestandard error of mean, asterisk indicates significant (p < .05) decrease in blast intensity.
Scale goes up to 8 as a maximum.
96
Table 3.4.Moderation models with bootstrapping method, using self-control training as predictor,mindfulness as moderator, blast intensities as outcome
Model testedNo
provocationLow
provocationHigh
provocation
B SE B SE B SE
Self-control training (c1) -.78 .55 -.81* .36 -.99* .39Mindfulness induction (c2) .18 .55 .004 .36 -.07 .39Self-control x mindfulness (c3) .52 1.09 .38 .72 -.57 .78R2 on interaction .23% .28% .51%Conditional effect of self-control trainingwithout mindfulness
-1.04 .77 -.99 .51 -1.27 .55
Conditional effect of self-control trainingwith mindfulness
-.53 .78 -.61 .51 -.71 .55
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
The moderation analysis was then repeated by controlling for the associated
personality variables (PA for all provocation levels, SP for provoked aggression; see
again Table 3.2). Controlling for the hypothesised covariates yielded a stronger main
effect of self-control training (Table 3.5), in which under provoked aggression, self-
control trained participants delivered lower levels of blast intensity than non self-
control trained participants. There was still no main effect of mindfulness, or
interaction effect of self-control training x mindfulness on blast intensity (Hypothesis
4). Higher SP predicted higher provoked aggression, but no differences in blast
intensity were found as a function of SP in the absence of provocation. Higher levels
of PA predicted more intense blasts across all provocation levels.
Table 3.5.Moderation models with bootstrapping method, using self-control training as predictor,mindfulness as moderator, blast intensities as outcome, controlling for covariates
Model testedNo provocation Low provocation High provocation
B SE B SE B SE
Self-control training (c1) -.71 .52 -.76* .33 -.96** .35
Mindfulness induction (c2) .45 .53 .26 .33 .19 .36
Self-control x mindfulness (c3) -.24 1.05 .11 .62 .28 .70
R2 on interaction .05% .02% .13%
Covariates:SP - - .65** .20 .89*** .22PANAS PA 1.10** .36 .94*** .23 .94*** .25Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
97
3.4.5.2 Maximum blast latency
During the TCRT, 83 participants out of 99 delivered maximum blasts to the
opponents. Censored survival analyses using Cox regressions were carried out to
assess the extent to which individuals were more likely to administer the maximum
blasts as a function of the mindfulness induction and self-control training conditions.
Contrary to prediction, our proposed model was not significant (2 = 1.63, p = .65).
The likelihood of delivering the maximum blast earlier was not predicted by the main
effect of mindfulness induction (Hypothesis 1b: Wald = 1.40, p = .24, 95%CI [77,
2.93], exp(B) = 1.50), self-control training (Hypothesis 2b: Wald = .16, p = .69, CI [.60,
2.15], exp(B) = 1.14), or the interaction between mindfulness induction and self-
control training (Hypothesis 3b: Wald = .31, p = .58, CI [.32, 1.88], exp(B) = .78).
The censored survival analyses were repeated to assess the maximum blast latency
as a function of the experimental conditions and the hypothesised covariates
(physical aggression). Our final model was still not significant (Hypothesis 4: 2 =
3.88, p = .42). The point at which participants delivered the maximum blast to
opponent was not influenced by the experimental conditions (p = .79, .37 and .67 for
the effect of self-control training, mindfulness induction, and self-control training x
mindfulness induction, respectively), or by the main effect of physical aggression
(Wald = 2.15, p = .14, CI [.93, 1.72], exp(B) = 1.26). Interestingly, when we analysed
this model separately for males and females, the findings varied (Figure 3.4).
For females (n = 44), the mindfulness induction x self-control training model was
significant (2 = 11.16, p < .05). There were significant main effects of mindfulness
induction (Wald = 6.66, p < .01, CI [1.41, 12.13], exp(B) = 4.13), and physical
aggression (Wald = 7.37, p < .01, CI [1.24, 3.87], exp(B) = 2.19) on maximum blast
latency. Specifically, females with no mindfulness induction and higher in physical
aggression were more likely to deliver the maximum blast. There was still no main
effect of self-control training (Wald = .83, p = .36, CI [.60, 4.08], exp(B) = 1.56), or
interaction effect of mindfulness induction x self-control training (Wald = 2.65, p =
.10, CI [.09, 1.25], exp(B) = .33) on maximum blast latency.
98
Figure 3.4. Sex differences in maximum blast latency. Error bars represent standard error ofmean, asterisk indicates that participants waited significantly longer (p < .01) before
delivering the maximum blast.
For males (n = 39), the mindfulness induction x self-control training model was not
significant (2 = 1.62, p = .81). Maximum blasts latency was not predicted by
mindfulness induction (Wald = .49, p = .48, CI [.23, 2.00], exp(B) = .68), self-control
training (Wald = 1.22, p = .27, CI [.21, 1.55], exp(B) = .57), interaction between
conditions (Wald = .36, p = .55, CI [.39, 5.74], exp(B) = 1.50), or physical aggression
(Wald = .001, p = .98, CI [.39, 5.74], exp(B) = .99).
3.4.5.3 Correlation between blast intensity and maximum blast latency
For participants who had delivered the maximum blast (N = 83), there were negative
correlations between maximum blast latency and blast intensity across levels of
provocation (r = -.40, p < .001 on no provocation, r = -.41, p < .001 on low
provocation, and r = -.23, p <.05 on high provocation). This finding indicated that
higher blast intensities were delivered by those who had delivered the maximum
blast to the opponent earlier. The correlation persisted for males under no
provocation and low provocation, and for females under low provocation only. For
males, the correlation between maximum blast latency and blast intensity became
non-significant under high provocation (r = -.30, p = .07). For females, the correlation
between maximum blast latency and blast intensity became non-significant under no
provocation (r = -.24, p = .12) and high provocation (r = -.19, p = .22).
99
3.4.6 Effect of mindfulness induction and self-control training on indirect
aggression
A bootstrapping procedure was employed again to analyse the proposed impact of
mindfulness induction and self-control training on each indirect aggression item (see
Table 3.6), but using 99% bias-corrected confidence intervals to survive Bonferroni
corrections (p < .01). The main impact of mindfulness induction on rating of the
opponent’s aggressiveness (Hypothesis 1c) did not survive Bonferroni corrections (p
< .03). There was no main effect of self-control training on indirect aggression
(Hypothesis 2c). Mindfulness induction did not moderate the link between self-
control training and indirect aggression (Hypothesis 3c).
Table 3.6.Moderation models with bootstrapping method, using self-control training as predictor,mindfulness as moderator, rating of opponent as outcome
Model testedAggressive Skilful Competitive Fair Play again
B SE B SE B SE B SE B SE
Self-control training (c1) .10 .25 -.08 .22 -.04 .22 -.15 .28 -.22 .31
Mindfulness induction(c2)
-.54* .25 -.002 .22 -.40 .22 -.43 .28 -.30 .31
Self-control xmindfulness (c3)
.20 .50 .48 .44 .24 .44 -.13 .57 .29 .61
R2 on interaction .15% 1.26% 30% .06% .22%
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
Because the indirect aggression ratings did not differ as a function of participant’ sex
(see section 3.4.3), we did not break down the moderation models by sex. As shown
in Table 3.7, no sex differences were found in ratings of indirect aggression.
Table 3.7.Sex differences based on rating of opponent
Indirect aggressionMales (n = 48) Females (n = 51)
tM (SD) M (SD)
Aggressive 4.27 (1.25) 4.00 (1.28) t(97) = 1.06, p = .29Skilful 5.06 (.98) 4.78 (1.17) t(95.08) = 1.30, p = .20Competitive 5.23 (1.08) 5.10 (1.12) t(97) = .60, p = .55Fair 5.02 (1.34) 5.04 (1.47) t(97) = -.07, p = .95Play again 5.48 (1.46) 5.16 (1.57) t(97) = 1.06, p = .29*p < .05;
**p < .01;
***p < .001
100
Based on the prior significant zero-order correlations between covariates and rating
indirect aggression (see again Table 3.2), the moderation analyses for ratings of the
opponent skill, competitiveness, and fairness were repeated by including the
hypothesised covariates (Hypothesis 4, see Table 3.8). There were no effects of
experimental conditions on ratings for these items. Those higher in curiosity rated
the opponent as being more skilful, whereas the negative impact of trait mindfulness
on this rating did not survive Bonferroni correction (p < .013). Those who were
higher in PA rated the opponent as being more competitive.
Table 3.8.Moderation models with bootstrapping method, using self-control training as predictor,mindfulness as moderator, rating of opponent as outcome, controlling for covariates
Model testedSkilful Competitive Fair
B SE B SE B SE
Self-control training (c1) -.07 .21 -.05 .21 -.12 .28Mindfulness induction (c2) -.05 .21 -.27 .21 -.34 .28Self-control x mindfulness (c3) .39 .41 .11 .41 -.22 .56R2 on interaction .83% .06% .15%Covariates:MAAS -.32* .13 - - - -TMS Curiosity .31** .11 - - - -PANAS PA - - .52*** .14 .35 .19Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
3.4.7 Correlation between blast intensity and indirect aggression
To provide converging evidence, a bivariate analysis was conducted on the
aggression measures. Participants evaluated opponents that they had blasted more
intensely as being less aggressive, but also more fair and a greater willingness to play
against the same opponent again (Table 3.9).
Table 3.9.
Bivariate correlations between blast intensity and rating of opponent
Indirect aggression No provocation Low provocation High provocation
Aggressive -.13 -.28** -.24*
Skilful .08 -.01 .02
Competitive .02 .00 .04
Fair .35** .42** .44**
101
Indirect aggression No provocation Low provocation High provocation
Play again .27** .21* .25*
*p < .05;
**p < .01;
***p < .001
3.4.8 Effect of mindfulness induction and self-control training on self-control
performance
To examine Hypothesis 5, we entered the hypothesised covariates into the
moderation models (i.e., sex, SP, and ethnicity; see again Table 3.2). There were no
significant main effects of mindfulness induction (c2: B = 9.83, SE = 15.66, p = .53) or
self-control training (c1: B = 3.62, SE = 15.22, p = .81) on handgrip power. Those who
received mindfulness induction did not show a better handgrip power than those
who did not receive mindfulness induction; and self-control trained participants did
not show a better handgrip power than non self-control trained participants.
Figure 3.5. Crossover interaction between mindfulness x self-control training on handgrip
power. Error bars represent standard error of mean, asterisk indicates significant (p < .05)
increase in handgrip power.
Contrary to prediction, handgrip power was influenced by a crossover interaction
between mindfulness induction and self-control training (c3: B = -64.59, SE = 30.46, p
< .05) with total variance uniquely attributable to the interaction of 4.07% (F(1, 90) =
4.50, p < .05). Specifically, a better handgrip power was shown by participants who
received one condition only (mindfulness induction or self-control training), but not
by those who received both conditions or did not receive any conditions (Figure 3.5).
102
A better handgrip power was also shown by males as opposed to females (B = 47.38,
SE = 15.59, p < .01). The partial effects of ethnicity (B = 5.31, SE = 12.79, p = .68) and
SP (B = -14.60, SE = 9.48, p = .13) on handgrip power were not significant.
The crossover interaction between self-control training and mindfulness induction
on handgrip performance diminished when males and females were analysed
separately. Specifically, handgrip power was no longer predicted by interaction
between mindfulness induction and self-control training, or by main effects of
mindfulness induction or self-control training (Table 3.10, Figure 3.6).
Table 3.10.
Moderation models with bootstrapping method, using self-control training as predictor,
mindfulness as moderator, handgrip power as outcome
Model testedMales (n = 48) Females (n = 51)
B SE B SE
Self-control training (c1) 31.94 24.96 -21.49 17.04
Mindfulness induction (c2) 5.62 24.98 11.26 17.04
Self-control x mindfulness (c3) -85.59 49.97 -37.13 34.09
R2 on interaction 6.04% 2.36%
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
Figure 3.6. Sex differences in handgrip power. Error bars represent standard error of mean.
103
3.5 DISCUSSION
3.5.1 Effect of mindfulness induction and self-control training on aggressive
behaviour
The current study experimentally activated individual’s levels of mindfulness
(mindfulness induction vs. relaxation) and self-control (self-control training vs. no
training), using 99 (51 females) university students. Contrary to prediction, the first
main analysis showed that mindfulness induction neither predicted direct aggression
(blast intensity administered to the opponent in the adapted TCRT task), nor
moderated the link between self-control training and direct aggression. This finding
runs counter the only reported aggression study in mindfulness (Heppner et al.,
2008). Specifically, Heppner et al. found that mindfulness induction reduced direct
provoked aggression amongst participants who had received rejection/insulting
feedback as opposed to those in rejection only and non-rejection conditions. It
seems unlikely that these contradictory findings would be attributable to the point at
which the mindfulness induction was given, as in Heppner et al.’s and our studies,
mindfulness induction was given before the participants received provocations.
Presumably, inducing a state of mindfulness may decrease the preparedness to
behave aggressively when aggressive impulses are subsequently stimulated.
It is possible that our mindfulness condition was not effective to produce any
significant effect on both direct provoked and unprovoked aggression (detailed in
latter part of this section). It is important to note, however, that participants
receiving mindfulness induction were no more aggressive than the control condition.
It has been speculated elsewhere that sustaining attention on the task at hand
without acting on the associated emotional contents might be especially difficult for
novice practitioner (Holzel et al., 2011; Sauer et al., 2013). Rather, this attempt may
lead them to over-regulate (e.g., suppress, avoid) their emotions and increase the
likelihood of aggressive behaviour (see Roberton et al., 2012). Despite the lack of
formal experience in mindfulness practices (only 5% participants reported currently
having mindfulness practice), initial encounter with this practice did not increase
their direct aggression.
104
It is also plausible that the impact of mindfulness induction was not readily observed
because our participants were, in fact, under the conditions of full self-control
resources. Research with respect to the strength model of self-control has typically
shown that experimental manipulations, such as glucose (Gailliot et al., 2007), self-
affirmation (Schmeichel & Vohs, 2009), and positive affect (Tice, Baumeister,
Shmueli, & Muraven, 2007), rarely increase self-control performance for non-
depleted participants. In the same way, it may be that mindfulness induction only
becomes effective under circumstances where individuals are depleted, independent
of provocation condition. The proposed impact will be tested under ego-depleted
condition in the next studies in this thesis.
On the other hand, blast intensity in response to provocations was significantly
decreased by a two-week training of improving posture. As a comparison, in Denson
et al.’s (2011) study, the effect of self-control training on provoked aggression was
shown only amongst those high in trait aggression; whereas in Finkel et al.’s (2009)
study, this effect was investigated under depleted condition (all participants were
asked to engage in an attention control task, before they responded to partner’s
provocation). We expanded these past studies by including a no-provocation
condition, showing that while training in self-control may increase self-control
resources, the benefits of this training on aggressive behaviour may be salient in
particular when aggressive impulses have been stimulated (i.e., following
provocation), even in the absence of depleted condition. This effect persisted after
accounting for significantly related personality variables (i.e., sensitivity to
provocation [SP] and positive affect), and was consistent across sexes. Thus
regardless of individual differences in sensitivity to provocation, mood, and sex,
increases in self-control resources may predict people’s ability to refrain from
behaving aggressively when the urge to aggress is relatively strong.
Resembling Lawrence and Hutchinson‘s (20013b) finding, individual differences in SP
also independently predicted direct provoked aggression. In addition, blast intensity
in all provocation levels was associated with post-TCRT positive affect. As the mood
measure was given after the TCRT task, this finding supports the view that those who
expressed their aggression in response to provocation may do so in order to improve
105
their mood (Bushman et al., 2001). However, in order to assess this fully, a baseline
mood assessment would have been needed. The subsequent studies reported in this
thesis will do just this.
The second main analysis revealed that the extent to which participants delivered
the maximum noise blast to the opponent was independent of mindfulness
induction and self-control training. However, when males and females were
analysed separately, a main effect of mindfulness induction occurred for female
participants. Specifically, the likelihood of delivering the maximum blast earlier
decreased for females with mindfulness induction, as opposed to females with no
mindfulness induction. This indicates that mindfulness induction may be a way to
reduce aggression in its more subtle/implicit form (as represents by the maximum
blast latency), and therefore more suitable to attenuate females’ than males’
aggression (for a discussion on various forms of aggression in the TCRT task, see
Giancola & Parrot, 2008).
In terms of the effect of the mindful and self-control manipulations on indirect
aggression, self-control training had no influence, whereas there was a minor
contribution of mindfulness induction. Specifically, participants receiving
mindfulness induction (compared with no induction) rated the opponent as being
less aggressive; however, this tendency did not survive Bonferroni correction. Whilst
it is difficult to raise any definitive conclusion based on the trivial effect of
mindfulness, one could argue that the lack of a role of self-control on indirect
aggression is due to the nature of the training used. That is, self-control training
involving physical behaviour may have a noticeable effect on direct (physical)
aggression but not on indirect aggression. Although training in one sphere of self-
control has been demonstrated to improve the general capacity of self-control (see
Baumeister et al., 2006; Hagger et al., 2010), future work could fruitfully explore
whether different types of training would produce differential effects on direct and
indirect aggression.
In addition, self-harm status had no influence on blast intensity or indirect
aggression, although the likelihood of delivering the maximum blast sooner was
106
approaching significance for self-harmers. As that self-harm status in the current
sample was related to most of the self-reported aggression measures (see section
3.4.2), it is plausible that under full self-control resource, the provocation procedure
in the TCRT task might also not be strong enough to activate the link between self-
harm and aggressive behaviour. Whether or not this link could be activated under
depleted self-control resource will be investigated in the next studies.
Finally, neither direct nor indirect aggression was not related to participant’s
background in current mindfulness practice. However, the unequal sample size
between experienced and non-experienced mindfulness practitioners might have
compromised this result. Likewise, direct and indirect aggression was not predicted
by ethnicity. Further cross-cultural comparison is indeed crucial (see Chapter 6), as
association between cultural background and different types of aggression has been
widely demonstrated (e.g., Forbes et al., 2009; Li et al., 2010; Magnis-Suseno, 1997).
3.5.2 Effect of mindfulness induction and self-control training on self-control
performance
Considering that males should be stronger than females with respect to their
handgrip power, the influence of participants’ sex was controlled in this analysis.
Mirroring the insignificant effect of mindfulness induction on blast intensity, no main
effect of mindfulness induction was found on self-control performance (i.e.,
handgrip power). Unusually, self-control performance was also not predicted by the
main effect of self-control training.
There was, however, a crossover interaction between mindfulness induction and
self-control training. Specifically, individuals receiving both mindfulness induction
and self-control training showed a lower handgrip power than those receiving one
condition only. As such, mindfulness and self-control appeared to work in an
antithetical way. Even though this possibility should be taken into account, it is
feasible that because the self-control performance was measured after the
experience of provocation, the expected independent effect of mindfulness
107
induction and self-control training as well as the complementary benefit of these
exercises could not be revealed. In the typical the strength model of self-control
studies, the effect of self-control training is measured on the performance in self-
control task only (see Baumeister et al., 2006) or on aggressive behaviour only
(Denson et al., 2011; Finkel et al., 2009); whereas in this study, self-control
performance was measured after we measured aggressive behaviour.
It is also plausible that the crossover interaction occurred because participants who
were at the same time experienced bolstered self-control resource (i.e., improved
capacity in self-control following the training) might have simply chose to let go of
any excessive controlled efforts (e.g., squeezing a handgrip for a lengthy period of
time). The mindful capacity also includes observing one’s own attempt to exert self-
control and to let go of any thoughts, emotions, or desires that are less meaningful
for oneself (see Brown et al., 2007a; Shapiro et al., 2006). Our next studies will test
whether this phenomenon also occur when participants lack of self-control resource.
3.5.3 Strengths and limitations of the current study
An important limitation of the current study is that we only managed to retain data
from 99 out of the optimum number of 128 participants (G*Power, 2013), mainly
due to the high attrition rate from Time 1 to Time 2 of the study.
Heppner et al. (2008) have provided preliminary support for benefit of brief
mindfulness inductions on direct provoked aggression. However, neither active
control condition nor manipulation check was included in their work. In the current
study, we found no significant differences between the mindfulness and relaxation
conditions in terms of state mindfulness, as assessed by the Toronto Mindfulness
Scale (TMS: Lau et al., 2006). Past study comparing a 1-month training in mindfulness
and somatic relaxation (i.e., progressive muscle relaxation, simple breathing
techniques, and guided imagery) showed that both training reduced distress and
improved positive mood, but only mindfulness training reduced distractive and
ruminative thoughts and behaviours (Jain et al., 2007). A longer period (7-week) of
108
both training also similarly increased well-being and reduced skin conductance
responses to unpleasant stimuli, but reductions in the emotional interference from
unpleasant stimuli were shown only by mindfulness trained participants (Ortner et
al., 2007). Given that in this study the mindfulness and relaxation exercises were
given only for 10 minutes, the time duration might not be strong enough to produce
any differential effects. As no past laboratory work has used mindfulness of walking
task before, we have no reference as to the immediate effect of this exercise, as well
as the accountability of the TMS to capture this effect. In our next studies, we will
employ a common method of laboratory inductions of mindfulness (i.e., mindfulness
of breathing exercise) and a neutral condition for the control group that have been
used in a recent study of mindfulness and ego-depletion by Friese et al. (2012).
We also cannot rule out the possibility that the efficacy of self-control training on
direct aggression is due to a non-equal involvement between the self-control
training and non self-control training participants. During the 2-week interim period,
self-control training participants were contacted on a daily basis via email, whereas
those in the no training group were only contacted immediately after completing the
online survey, a week before the experiment, and approaching the day of the
experiment. To keep experimenter contact equal, future work should incorporate a
dummy communication for the control condition.
One limitation of the handgrip measure is that we did not assess the baseline
duration for handgrip power, because we wanted to avoid the occurrence of ego-
depletion effect. Given that that this task is sensitive to individual differences in
physical hand strength (Muraven et al., 1999), participant’ sex was assessed as an
additional factor. Handgrip power may also be influenced by individual differences
not measured in the current study. In our next studies, we will include the measure
of baseline handgrip power, and used a manual hand exerciser instead of a digital
one to allow a more precise determination of when the handgrip was released.
Despite the limitations, this study has provided a direct comparison for the efficacy
of mindfulness and self-control interventions on reducing aggressive behaviour using
109
different types of aggression. Additionally, a behavioural measure of self-control
performance as additional dependent variable was also entailed.
3.5.4 Conclusions and next steps
Findings from the current study reveal that reductions in aggressive behaviour could
occur by experimentally manipulating individual’s levels of mindfulness and self-
control; however, this effect may vary for different types of aggression. Reduction in
blast intensity following provocations was related to training in self-control, whereas
mindfulness induction only reduced the more implicit form of aggression (i.e., the
maximum blast latency for female participants). While the complementary effect of
mindfulness and self-control was not shown on aggressive behaviour, a reverse
effect was shown on self-control performance (i.e., handgrip power). Provided that
the inconsistent role of mindfulness induction over self-control training is likely to
occur due to participants’ sufficient resources to perform self-controlled acts, the
next studies will examine the effect of mindfulness induction on aggressive
behaviour and self-control performance under ego-depletion condition.
110
CHAPTER FOUR
Study 3: Effect of mindfulness induction on the aggressive behaviour following
depletion (British sample)
4.1 INTRODUCTION
In Study 2 (Chapter 3), the effect of a brief mindfulness induction on aggressive
behaviour in the adapted TCRT (Taylor, 1967) task occurred only in the implicit form
of aggression (i.e., the delay duration before the maximum aggressive response was
delivered, particularly for females). By contrast, direct provoked aggression was
effectively reduced by training in self-control. Individuals who performed both
mindfulness induction and self-control training showed lower self-control
performance (i.e., handgrip power) compared to those who performed one exercise
only. Whether or not this inconclusive pattern of results was due to the individuals
having sufficient resource to exert self-control is examined in this study. Specifically,
we examine the moderating effect of mindfulness induction on the link between
ego-depletion and aggressive behaviour and self-control performance. Importantly,
to clarify the role of cultures on aggression, only native British individuals are
recruited in the current study. The roles of individual differences in self-harm,
sensitivity to provocations (SP) and frustrations (SF) are also assessed.
4.1.1 Self-control ego-depletion and aggressive behaviour
According to the strength model proposed by Baumeister and colleagues,
performing self-control in one sphere depletes the common “resource” of self-
control, resembling a muscle that gets tired after exertions (see Hagger et al.’s meta-
analysis, 2010). This temporary state of diminished self-control strength is known as
ego depletion. Under depleted state, people’s ability to perform further, seemingly
unrelated self-control acts is compromised.
111
In laboratory settings, ego-depletion is typically induced by the dual-task self-control
procedure, using various spheres of self-control such as controlling attention,
controlling emotions, controlling impulses, controlling thoughts, cognitive
processing, choice and volition, and social processing (Baumeister, Vohs, et al.,
2007). Participants in the depletion group are asked to perform two consecutive
tasks requiring self-control, each task commonly lasts for 3 until 10 minutes (see
Hagger et al., 2010). The first task serves as depleting task, the second one serves as
dependent task. Control participants are also asked to perform two consecutive
tasks, in which the first task is usually a less effortful version of the depleting task (so
that participants can enact their impulsive responses), the latter task is the same
dependent task as the depletion group. The strength model predicts that depleted
participants would show impaired performance on the dependent task relative to
control participants.
In the area of aggression, Stucke and Baumeister (2006) found that compared to
non-depleted participants, depleted participants who resisted eating tempting food
or stifled their physical and emotional reactions delivered more negative ratings to
an experimenter that had previously insulted them, independent of mood, anger, or
frustration due to depletion condition. De Wall et al. (2007) showed that higher
aggressive responses in the TCRT task (as measured by the composite score of blast
intensity and duration in the first unprovoked trial) or in candidate job evaluation
were displayed by depleted participants only when they had been insulted, but not
in the absence of insulting provocation. Similarly, depletion was also shown to
increase aggression toward intimate partners only when partners had ostensibly
provoked participants with negative feedback (Finkel et al., 2009). Altogether, these
findings point out that the effect of depleting self-control resources may only
manifest in aggressive behaviour if people encounter subsequent aggression-
triggering events. In the same way, our result in Study 2 (Chapter 3) showed that the
effect of bolstering self-control resources (i.e., via self-control training) in the
reductions of aggressive behaviour only occurred in the presence of provocation.
112
Inconsistent role of provocations, however, was shown in Vohs et al.’s (2011) study
using manipulations of depletion (emotion regulation while watching disgusting
movie clip) and fatigue conditions (24 hours of total sleep deprivation). Rather than
including insult as a method of provocation, the opponent’s choice of blast intensity
in the TCRT task (displayed as one unit higher than the participant set for the
opponent) was used as a sole procedure to provoke aggression. Increases in blast
intensity for depleted participants were found not only when they had received
provocation from the opponent in the losing trials, but also during the initial
unprovoked trial.
Whilst Vohs et al.’s work did not specifically address the role of provocation, it
indicates that depletion may directly increase aggressive behaviour in the absence of
provocation. Therefore, as in Study 2, the current study distinguishes blast intensity
across three provocation levels in the TCRT task (i.e., no provocation, low
provocation, and high provocation) and investigates whether depleting self-control
capacity could increase aggressive behaviour. In this way, we can examine the effect
of depletion on aggressive behaviour across provocation levels.
4.1.2 Role of mindfulness induction in counteracting the effect of depletion on
aggressive behaviour
Outside the aggression literature, several variables (e.g., glucose, humour, laughter,
and other positive emotions, cash incentives, implementation intentions, social
goals) have been found to moderate the depletion effect on performance on self-
control task (see Baumeister, Vohs, et al., 2007). However, none of these variables
completely counteracted the depleted resource, and conversely may “all operate by
inducing the person to expend more of the depleted resource” (Baumeister, Vohs, et
al., 2007, p.353). In the current study, we examine whether a brief period of
mindfulness induction may be a further way to counteract the effect of depletion on
aggressive behaviour and self-control performance.
113
Of particular relevance is Friese, Messner and Schaffner’s (2012) experiment using a
5 min period of mindfulness meditation. Sixty-six participants, approached at the
end of the second day of a 3-day introductory seminar on mindfulness meditation in
German and Swiss cities, were assigned to one of three conditions: emotion
suppression (induced by disgusting movie clip), no emotion suppression (in which
participants were asked to watch the same movie clip while allowing all emotions
that may arise), and emotion suppression plus mindfulness meditation. As predicted,
participants who had meditated after emotion suppression task performed better on
subsequent self-control task (i.e., the crossing-out-letters task to discriminate
between adjacent and similarly looking, but slightly different letters) than those in
the suppression condition and equally well as those not suppressing emotion. The
current study aims to test whether the moderating effect of mindfulness induction
following depletion also applies on aggressive behaviour. We extend Friese et al.’s
(2012) work via a fully crossed between subjects design (mindfulness x depletion).
Friese et al. (2012) suggested that the benefit of mindfulness on self-control
performance following depletion may be attributable to increase self-awareness and
a feeling of deep relaxation associated with mindfulness. As presented in Chapter 1,
a depletion effect on aggressive behaviour may also be attenuated by the mindful
reduction in emotional reactivity to aversive or emotionally provocative stimuli. In
the current study, this proposed effect is tested under provoked and unprovoked
aggression in the TCRT task. As in Study 2 (Chapter 3), we assess the maximum blast
latency (the number of trials participants waited before delivering the maximum
blast) as a measure of how long individuals can inhibit a desire to retaliate at
maximum level, and participants’ anonymous reputation damage of the opponent to
provide alternative ways of aggression measures (see Giancola & Parrot, 2008).
In addition, to assess the broader influence of mindfulness on self-control, we
measure the effect of mindfulness induction on self-control performance (i.e.,
handgrip task) at two time points: prior to the provocation procedure and after the
114
provocation procedure. In line with Study 1 and 2, we include measures of individual
differences in mindfulness, self-control, aggression, self-harm, SP and SF.
4.2 HYPOTHESES
The current study predicts that:
1. Compared to ego-depleted participants, those non ego-depleted will:
a. Deliver lower levels of blast intensity to the opponent, particularly on low
provocation trials. No strong prediction is made regarding the levels of
blasts delivered under no provocation or high provocation trials.
b. Wait longer before delivering the maximum blasts to the opponent.
c. Deliver lower ratings of anonymous reputation damage to the opponent.
2. Mindfulness induction will moderate the effect of ego-depletion on aggressive
behaviour. Specifically, compared to depleted participants without mindfulness
induction, depleted participants who receive mindfulness induction will:
a. Deliver lower levels of blast intensity to the opponent particularly on low
provocation trials. No strong prediction is made under no provocation or
high provocation trials.
b. Wait longer before delivering the maximum blasts to the opponent.
c. Deliver lower ratings of anonymous reputation damage to the opponent.
3. The moderation of mindfulness on the ego-depletion and aggressive behaviour
link will persist after accounting for the potential influence of individual
differences in mindfulness, self-control, aggression (i.e., trait aggression,
sensitivity to provocations [SP] and frustrations [SF]), sex, and self-harm.
In addition, we predict that:
4. Mindfulness induction will moderate the effect of ego-depletion on self-control
performance, as measured by the changes in handgrip stamina after accounting
for covariates:
115
a. Prior to the provocation procedure
b. After the provocation procedure
Given that the effect of mindfulness induction may not be salient when individuals
have sufficient self-control resources (i.e., not ego depleted, See Chapter 3), no
prediction is made for the main effect of mindfulness induction on aggressive
behaviour and self-control performance.
Altogether, we hypothesise that that ego-depletion will predict higher levels of
aggression and impaired self-control (handgrip) performance amongst participants
not receiving mindfulness induction but not amongst participants receiving
mindfulness induction, after accounting for covariates (personality variables).
4.3 METHODS
4.3.1 Participants
A total of 118 native British participants (110 from the University of Nottingham
[UoN] and 8 from Nottingham Trent University [NTU]) participated in the study.
Based on sex, participants were randomly assigned to one of four group (see details
in the next section). Eight participants were discarded from final analysis due to:
expressing spontaneous suspicions regarding the TCRT task (5 participants), data
collection stopped prematurely due to computer error while performing the TCRT
program (1 participant), and expressing confusion with the mindfulness induction
procedure after they had performed the task (2 participants), resulting in 110
participants (58 females). Participants’ mean age was 19.52 (SD = 2.03). Concerning
current mindfulness practice, 92.7% participants reported never having encountered
mindfulness, 6.4% practised them once a year, 1% (1 participant) on monthly basis,
and none practised them on weekly or daily basis. The experiment lasted from
October until November 2012.
116
4.3.2 Design
We used a fully-crossed 2 (ego-depletion task vs. no depletion) x 2 (mindfulness
induction vs. no induction) between subjects experimental design in the current
study. A stratified sampling technique was also employed to control the effect of sex
(see Figure 4.1 for flow of participants).
Figure 4.1. Flow of participants through Study 3.Note. M = Males; F = Females
4.3.3 Trait and state measures
The self-reported measures in Study 1 and 2 (Appendix 2.1) were employed to assess
individual differences in mindfulness (Mindfulness Attention Awareness Scale
[MAAS]: Brown & Ryan, 2003), self-control (Self-Control Scale [SCS]: Tangney et al.,
117
2004), aggression (Aggression Questionnaire [AQ]: Buss & Perry, 1992), sensitivity to
frustrations and provocations (Situational Triggers of Aggressive Responses [STAR]
scale: Lawrence, 2006), and self-harm (Deliberate Self-Harm Inventory [DSHI]: Gratz,
2001). See Section 2.3.3 (Chapter 2) for the full description of the trait measures.
As in Study 2, we replaced the measure of anxiety and depression (Hospital Anxiety
and Depression [HADS]: Zigmond & Snaith, 1983) from Study 1 with the mood
measure (Positive Affect, Negative Affect Schedule [PANAS]: Watson et al, 1988).
The PANAS was given twice: following depletion task and following the TCRT task.
The efficacy of the mindfulness induction was assessed using Toronto Mindfulness
Scale (TMS: Lau et al., 2006), compromising mindfulness state of curiosity and
decentering. Description of the TMS is presented in section 3.3.3, Chapter 3 (see
Appendix 3.1 for the PANAS and TMS).
The efficacy of the depletion manipulation was assessed with two following
questions (i) “How difficult was it for you to follow the instructions during the video
clip”, and (ii) “How much you had to control your attention to follow the instructions
during the video clip”, on a 7 point Likert-scale (1 = not at all and 7 = very much). A
higher score would indicate higher levels of depletion.
4.3.4 Experimental manipulations
4.3.4.1 Ego-depletion task
Ego-depletion was induced using the dual-task self-control procedure (see section
4.1.1), in which participants in the depletion group were asked to perform two
consecutive tasks requiring self-control, whereas participants in the control
condition (no depletion) were also asked to perform two consecutive tasks, but only
the second task required self-control. For the first self-control task, we used an
attention control procedure that has been used in previous aggression experiments
(e.g., DeWall et al., 2007; Finkel et al., 2009). Specifically, all participants were asked
118
to watch a 6-min videotape (without audio) depicting a woman being interviewed by
an off-camera interviewer, and were told that they would later be making person
perception judgments of the interviewee. In addition to the women being
interviewed, a series of common one-syllable words (e.g., glue, tire, book) appeared
at the bottom of the screen for 10 s each. Participants in the depletion condition
were instructed “not to read or look at any words that may appear on the screen”,
and to redirect their gaze immediately if they caught themselves looking at the
words instead of the woman’s face. Participants in the no depletion condition were
not given any instructions and were not told in advance that there would be words
at the bottom of the screen.
For the second self-control task, we used performance in a hand-grip task (discussed
in the latter part of this section).
4.3.4.2 Mindfulness induction
For the mindfulness induction, we used Mark Williams’ audio instruction for
“Mindfulness of body and breath” (Williams & Penman, 2011; see Appendix 4.1),
typically used in the MBCT course (Oxford Mindfulness Centre, 2002; see also
Kramer, Weger, & Sharma, 2013). Participants were guided with an audio
commentary to direct their attention towards witnessing the full sensations of
breathing and to observe them without the intention of altering them. They were
also told to notice in an accepting manner when their minds wander and gently
return their focus to the sensations of breathing. Variants of these instructions were
repeated every 60 s for 15 minutes.
For the control condition, we used a neutral educational information. Participants
listened to two educational excerpts and completed a word search puzzle for a total
15 minutes. A similar procedure has been used elsewhere (Erisman & Roemer, 2010)
as a control condition for the effect of mindfulness condition on emotion regulation.
119
4.3.4.3 Direct aggressive behaviour
Participants’ direct aggressive behaviour was assessed using the adapted Taylor
Competitive Reaction Time (TCRT: Taylor, 1967; for a complete description see
section 3.3.4.3, Chapter 3). The TCRT task has been employed in studies measuring
the effect of depletion (e.g., De Wall et al., 2007; Finkel et al., 2009; Vohs et al.,
2011) and mindfulness (Heppner et al., 2008) on aggressive behaviour.
As in Study 2, participants experienced no provocation, low provocation, and high
provocation trials. The blast intensities delivered to the opponent under these three
trials were used as a measure of direct aggression. The maximum blast latency (the
number of trials participants waited before delivering the maximum blast) was also
used as an additional measure of direct aggression.
4.3.4.4 Indirect aggression
As in Study 2, indirect aggression was assessed with five ratings of the opponent’s
reputation damage (i.e., opponent’s aggressiveness, skill, competitiveness, and
fairness, and participants’ willingness to play with the same opponent again in future
paid studies) on a 7-point scale (1 = completely disagree and 7 = completely agree).
4.3.4.5 Self-control performance
A handgrip task was used to measure self-control performance following depletion
(see Hagger et al., 2010). To allow a more precise determination of when the
handgrip was released, we replaced the digital handgrip in Study 2 with a manual
hand exerciser (consisting of two handles and a metal spring) and inserted a small
wad of paper between the far end of the handles. Self-control performance was
measured by the duration of time the participant was able to continuously squeeze
the handgrip using the dominant hand until the wad fell out.
120
Each participant was asked to perform the handgrip task three times (i) at the
beginning of the study (baseline handgrip stamina), (ii) immediately following the
mindfulness induction (as a measure of performance in the second self-control task),
and (iii) at the completion of the study (to test whether the effect of mindfulness
induction would persist). Self-control performance was calculated twice as the
changes in handgrip duration compared with baseline duration.
4.3.5 Power calculation
A similar previous study (Friese et al., 2012) of the effect of mindfulness induction on
self-control performance following ego-depletion assigned 66 participants to one of
three conditions (i.e., no depletion, depletion, and depletion plus mindfulness). Thus
an additional 22 participants (88 in total) would be necessary for four conditions in
the current study. When calculated with G*Power 3 (2013), a total 128 participants
is required to detect a medium effect (d = .25) for four group and 1 covariate (sex),
and given the power of .80 and an alpha level of .05. Based on these considerations,
the current study would require between 22 to 32 participants (11 to 16 males or
females) for each group.
4.3.6 Procedure
Potential participants from University of Nottingham (UoN) were contacted via
posters/leaflets on the university campus and departmental mailing lists (Appendix
4.2). Those from Nottingham Trent University [NTU]) were recruited by asking the
researcher’s colleagues (NTU students) to help distribute the leaflets around their
campus. Participants were told the study investigated the effects of personality traits
and experimental treatments on the way people perform in a competitive reaction-
time task. To be able to participate, they have to speak British English as their first
language. They were given a £5 inconvenience allowance and a chance to win £50
121
incentive for the fastest participants’ reaction-time across the entire study.
Alternatively, experimental credit points were given for UoN Psychology students
signing up via the Research Participation Scheme.
Participants came into the lab in the School of Pychology. At first, information about
the study was presented (Appendix 3.8). After giving informed consent (Appendix
3.9), they were asked to provide demographic information (i.e., sex, age, current
mindfulness practice) and completed an online questionnaire of the trait measures
(Appendix 2.1). Immediately after the trait measures, participants were asked to
furnish a baseline handgrip measure. The experimenter stopped the stopwatch as
soon as the wad inserted between the grip handles fell out.
Next, all participants were asked to watch the 6 min attention control videotape.
Participants were left alone to do this task and were asked to open the cubicle when
they had finished watching the video clip. The experimenter then returned to the
cubicle and asked participants to complete two questions measuring the
effectiveness of the depletion procedure, three dummy questions about the person
perception judgments of the interviewee on the videotape, and the first PANAS
(Watson et al., 1988),
Half of the participants in the ego-depletion group and half of those in the no-
depletion condition were then given a mindfulness induction task, while the rest of
the participants were given the neutral education information. Then the
experimenter left the room again, while participants listened to and practised the
appropriate audio guided instruction for 15 minutes. When the participant opened
the cubicle door, the experimenter entered the cubicle and asked participants to
complete the TMS (Lau et al., 2006), followed by a handgrip task.
Next, the participants were told that the experimenter needed to prepare their TCRT
opponent. After leaving the room for a while (ostensibly to prepare the opponent),
the experimenter returned. Participants were then left alone to perform the TCRT
task. When the participant opened the cubicle door to indicate the end of the TCRT
122
task, the experimenter asked the participants to complete the second PANAS and
perform a final measure of handgrip stamina. Finally participants were debriefed and
given help and support information (Appendix 3.10).
4.3.7 Data analysis
Statistical analyses were performed on the data using IBM SPS statistics 20.0 (2011).
First, associations between personality variables, and between personality variables
and the experimental outcomes (i.e., aggressive behaviour and self-control
performance) were examined using zero-order correlations or independent t-tests
(for sex and self-harm). The efficacy of the experimental conditions was assessed
using independent t-tests or one-way ANOVAs.
4.3.7.1 Moderation of mindfulness induction on the link between depletion and
aggressive behaviour
For the primary hypotheses, we used self-control ego-depletion as the predictor,
mindfulness induction as the moderator, and aggressive behaviour as the outcome.
Aggressive behaviour was measured in terms of (i) blast intensity on no provocation,
low provocation, high provocation (Hypothesis 2a), (ii) maximum blast latency
(Hypothesis 2b), and (iii) indirect aggression ratings/damage to opponent’s
reputation (Hypothesis 2c). For Hypothesis 2a and 2c, the moderating effect of
mindfulness induction on the link between self-control ego depletion and aggressive
behaviour is tested with a bootstrapping method, using model “number 1” in the A.
F. Hayes’s (2012a) PROCESS macro for simple moderation (explanation for this
model can be found in section 2.3.6, Chapter 2). Hypotheses 2a and 2c would be
supported if the interaction coefficient between depletion and mindfulness
induction is statistically different from zero. Specifically for Hypothesis 2c (five items
123
of indirect aggression), we tested if the effect would survive Bonferroni corrections
set at p <.05 /5 items, or p < .01.
Hypothesis 2b would be supported if the interaction between mindfulness and
depletion significantly predicts the maximum blast latency (i.e., the number of trials
participants waited before delivering the maximum blast). To test this prediction, we
conducted a censored survival analysis using Cox regressions to assess the extent to
which participants would be more likely to administer the maximum blast to the
opponent as a function of experimental conditions.
Hypothesis 1a, 1b, and 1c would be supported if there is a significant main effect of
depletion on blast intensity, maximum blast latency, and ratings of the opponent.
To provide support for Hypothesis 3, the moderation models were repeated while
accounting for the influence of the hypothesised covariates based on the prior
significant zero-order correlations between covariates and aggressive behaviour.
4.3.7.2 Moderation of mindfulness induction on the link between depletion and
self-control performance
The moderating role of mindfulness induction on the link between ego-depletion
and self-control performance were also examined using bootstrapping method. Self-
control ego-depletion was used again as the predictor, mindfulness induction as the
moderator, and self-control performance as the outcome. Specifically, self-control
performance was measured in terms of (i) Change 1 (changes in handgrip duration
prior to the TCRT task relative to baseline), and (ii) Change 2 (post-TCRT task
handgrip duration relative to baseline). We expected that the interaction coefficient
between depletion and mindfulness induction on Change 1 (Hypothesis 4a) and
Change 2 (Hypothesis 4b) would be statistically different from zero after accounting
for the potential influence of the hypothesised covariates, based on the prior
significant zero-order correlations between covariates and self-control performance.
124
4.4 RESULTS
4.4.1 Baseline levels
To ensure that our randomisation procedure was successful, the means of trait
measures were compared for each group (i.e., depletion plus mindfulness, depletion
no mindfulness, no depletion mindfulness, no depletion no mindfulness). The group
were not significantly different from each other in any of the self-reported trait
measures p =.13 to .96). Due to the small number of participants who reported
having had experience with mindfulness practice (only 1 participant currently
practised mindfulness on monthly basis), we did not carry out any comparisons
based on this factor.
4.4.2 Preliminary analysis
Table 4.1.Zero order correlations and psychometric properties of measures
Measures 1 2 3 4 5 6 7 8 9
MAAS (1) 1.00SCS (2) .47
***1.00
Total AQ (3) -.38***
-.47***
1.00AQ Physical (4) -.17 -.24
*.69
***1.00
AQ Verbal (5) -.17 -.30**
.75***
.31**
1.00AQ Anger (6) -.37
***-.42
***.76
***.40
***.45
***1.00
AQ Hostility (7) -.43***
-.43***
.70***
.26**
.39***
.42***
1.00STAR Provocations (8) -.26
**-.31
**.59
***.35
***.36
***.52
***.50
***1.00
STAR Frustrations (9) -.32**
-.21*
.46***
.14 .27**
.55***
.43***
.76***
1.00
M 3.72 3.05 2.43 2.04 2.88 2.22 2.59 2.88 2.54SD .67 .56 .57 .83 .84 .72 .79 .78 .76
Cronbach .84 .77 .89 .87 .77 79 .78 .89 .86
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression
Questionnaire; STAR = Situational Triggers of Aggressive Responses*p < .05;
**p < .01;
***p < .001
Table 4.1 presents the zero-order correlations amongst trait measures and
descriptive statistics (N = 110). Internal reliabilities of measures were good ( = .77
to .89). Trait mindfulness and trait self-control were positively correlated to each
125
other and negatively correlated to most of the aggression measures, replicating the
findings of Study 1 and 2.
A series of independent samples t-test revealed some significant sex differences in
trait variables. Compared to females (n = 58), males (n = 52) scored significantly
higher on physical aggression (t(9.34) = 3.21, p < .01; M males = 2.30, SD = .77 vs. M
females = 1.82, SD = .65) and significantly lower on sensitivity to frustrations (SF:
t(108) = -2.45, p <.05; M males = 2.36, SD= .76 vs. M females = 2.71, SD = .72). No
significant differences were found between self-harmers (n = 11) and non self-
harmers (n = 99) in any of the trait measures (p = .10 to .92). None of the
associations were significantly altered when we broke down the effect based on sex
(r = -.18 to .80) or self-harm (r = .13 to .77).
4.4.3 Zero-order correlations between personality variables and experimental
outcomes
Table 4.2 shows the associations between personality variables and (i) direct
aggression: blast intensity on each provocation level and maximum blast latency, (ii)
indirect aggression, and (iii) self-control performance/handgrip stamina: Change 1
(changes in pre-TCRT task handgrip duration), and Change 2 (changes in post-TCRT
task handgrip duration). Specifically for testing Hypothesis 3 and 4, significantly
associated personality variables would be controlled in further analyses.
4.4.3.1 Associations with direct aggression
Resembling the findings from Study 2, sensitivity to provocations (SP) was positively
associated to provoked direct aggression, but not to unprovoked direct aggression.
Males responded with higher blast intensity than females following low provocation,
but not under high provocation; whereas sex differences on no provocation were
126
approaching significant (see Table 4.3). Direct aggression did not differ as a function
of self-harm status (p = .18 to .90).
Examining those who delivered a maximum blast during the TCRT, those higher in
physical aggression delivered the maximum blasts to the opponent sooner.
Interestingly, females (n = 36) relative to males (n = 41), also delivered the maximum
blast earlier in the TCRT (t(75) = 3.59, p < .001; M males = 8.95, SD = 8.13 vs. M
females = 16.56, SD = 10.43). No differences were found between self-harmers (n =
8) and non self-harmers (n = 69; t(75) = .30, p = .77) on maximum blast latency.
4.4.3.2 Associations with indirect aggression
It can be seen from Table 4.2 that traits mindfulness and self-control were only
negatively associated to the rating of opponent fairness. Rating of this item was also
related, positively, to sensitivity to provocations and frustrations. Self-harmers also
rated the opponent as significantly more fair than non self-harmers (t(19) = 2.45, p <
.05; M self-harmers = 5.27, SD = .91 vs. M non self-harmers = 4.48, SD = 1.69).
Negative affect was negatively correlated with ratings of opponent’s skilfulness and
of participants’ willingness to take a part in against the same opponent again. No sex
differences (p = .06 to 50) were found in any ratings of indirect aggression.
4.4.3.3 Associations with self-control performance
Self-control performance (i.e., handgrip stamina) was negatively associated with
physical aggression. Females were more able to maintain their baseline handgrip
stamina than males in Change 1 (t(79.06) = 3.84, p < .0001; M male = -16.25, SD =
22.49 vs. M female = -3.02, SD = 12.83) and Change 2 (t(73.31) = -2.43, p < .05; M
male = -12.87, SD = 22.34 vs. M female = -4.54, SD = 11.22). No significant
differences between self-harmers and non self-harmers in handgrip stamina was
observed in Change 1 (t(108) = -1.20, p = .23) or Change 2 (t(108) = -.08, p = .94).
127
Table 4.2.Zero order correlations of personality measures, aggressive behaviour, and self-control performance
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression Questionnaire; STAR = Situational Triggers of Aggressive Responses;TMS = Toronto Mindfulness Scale, PANAS 1 = Positive Affect Negative Affect following depletion; PANAS 2 = Positive Affect Negative Affect post-TCRT taskAll correlations were based on N = 110, except for maximum blast latency. The correlations for maximum blast latency was made only for participants whodelivered a maximum blast (N = 77).*p < .05;
**p < .01;
***p < .001
128
4.4.4 Manipulation checks
The depletion manipulation was partly effective. Participants in the attention control
(depletion) condition (n = 55) reported having controlled their attention to a greater
extent than no control participants (n = 55; t(108) = 2.26, p < .05; M attention control
= 4.91, SD = 1.78 vs. M no control = 4.20, SD = 1.50), but did not rate the task as
more difficult than those in the no-control condition (t(108) = -.42, p = .68; M
attention control = 3.31, SD = 1.95 vs. M no control = 3.45, SD = 1.69).
Participants in the mindfulness breathing (n = 55) reported higher decentering than
no induction participants (n = 55; t(108) = 2.65, p < .05; M mindfulness induction =
3.03, SD = .60 vs. M no induction = 2.70, SD = .70), but did not show higher levels of
Curiosity (t(117) = 1.15, p = .25; M mindfulness induction = 3.50, SD = .81 vs. M no
induction = 3.31, SD = .88). Both mindfulness subscales shown good internal
consistency ( = .84 and .71 for curiosity and decentering, respectively) and were
positively correlated (r = .58, p < .0001).
Following the attention control task, participants in the depletion condition reported
higher PA than non-depleted ones (t(108) = 2.24, p < .05; M attention control = 2.66,
SD =.68 vs. M no control = 2.37, SD = .70), but not higher NA (t(108) = .57, p = .57; M
attention control = 1.26, SD =.38 vs. M no control = 1.23, SD = .24). Both mood
subscales immediately following the attention control task showed good internal
consistency ( = .87 and .72 for PA and NA, respectively) but were not significantly
correlated (r = .14, p = .15).
It should be noted that ego-depletion studies generally demonstrate no relationship
between depletion and mood, although at times there is an increase in negative
affect due to the aversiveness of depleting tasks (see Hagger et al., 2010). In
contrast, the current participants reported significant postdepletion increases in PA,
indicating that they might have seen the task as being more interesting. Importantly,
however, PA was not correlated to aggression (see Table 4.2, section 4.4.3.1). So any
129
differences in the experimental outcomes would not be related to differences in
participant’s mood following depletion.
Both mood subscales also showed good internal consistency ( = .91 and .77 for PA
and NA, respectively) and were positively correlated (r = .63, p < .0001) when given
following the TCRT task. No differences were found between the four between
subjects group on PA (F(3,106) = .63, p = .60) or NA (F(3,106) = .45, p = .72).
4.4.5 Moderation of mindfulness induction on the link between depletion and
direct aggression
4.4.5.1 Levels of blast intensity
Figure 4.2. Effect of level of provocations on blast intensity.Error bars indicate standard error of mean. Scale goes up to 8 as a maximum.
As typically found in studies varying provocation from the opponent (e.g., Anderson
et al., 2008; Lawrence & Hutchinson, 2013a, 2013b; see also Study 2 in Chapter 3),
intensity was significantly affected by level of provocations (F(1.31, 142.94) = 42.15,
p < .001; Figure 4.2). Specifically, participants delivered more intense blasts on high
provocations (M = 4.14, SD = 2.17) compared to low provocations (M = 3.38, SD =
130
2.00; p < .0001), and on low provocation compared to no provocation (M = 2.61, SD
= 2.50; p < .0001).
A comparable pattern of results was revealed when we performed a 2 (sex) by 3
(provocation level) mixed analysis ANOVA, in which blast intensity was still affected
by level of provocations (F(1.31, 141.01 = 41.40, p < .001), but not by its interaction
with sex (F(1.31, 141.01) = 1.09, p = .32). The main effect of sex is shown in Table 4.3,
in which males delivered a significantly higher blast than females following low
provocation. Sex differences on no provocation were approaching significance.
Table 4.3.Sex differences on blast intensity on each provocation level
Level of provocationsMales (n = 52) Females (n = 58)
tM (SD) M (SD)
No provocation 3.08 (2.89) 2.21 (2. 03) t(90.31) = 1.80, p = .08
Low provocation 3.82 (2.24) 3.00 (1.69) t(94.26) = 2.13*
High provocation 4.35 (2.29) 3.94 (2.06) t(108) = .98, p = .33*p < .05;
**p < .01;
***p < .001
The moderation model was tested across three provocation levels, using a
bootstrapping method, with 95% bias-corrected confidence intervals (based on
5,000 bootstrap resamples, N = 110). All predictors were mean-centered. Results in
Table 4.4 depict that Hypothesis 2a was partially supported such that there was
significant effect of mindfulness induction on the link between depletion and blast
intensity under no provocation and low provocation, but not under high
provocations. In other words, under no provocation and low provocation, depleted
participants who received mindfulness induction delivered lower levels of blast
intensity than depleted ones with no induction; whereas non-depleted participants
with mindfulness induction displayed comparable blast intensity with non-depleted
ones with no induction (See Figure 4.3). Additionally, depletion (Hypothesis 1a) and
mindfulness influenced unprovoked, but not provoked direct aggression. Specifically,
131
those who were depleted or did not received mindfulness induction delivered higher
levels of blast intensity under no provocation only.
Table 4.4.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, blast intensity as outcome
Model testedNo
provocationLow
provocationHigh
provocation
B SE B SE B SE
Depletion (c1) 1.24** .45 .62 .37 .42 .41Mindfulness induction (c2) -.95* .45 -.62 .37 -.62 .41Depletion x mindfulness (c3) -1.90* .90 -1.55* .74 -1.25 .82R2 on interaction 3.63%* 3.79%* 2.07%Conditional effect of depletion withoutmindfulness
2.19*** .64 1.40** .52
Conditional effect of depletion withmindfulness
.29 .64 -.15 .52
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
Figure 4.3. Mindfulness × depletion effect on blast intensity across provocation levels.Error bars indicate standard error of mean, asterisk indicates significant (p < .05) decrease in
blast intensity. Scale goes up to 8 as a maximum.
132
As no personality variables were significantly associated with blast intensity under no
provocation (see Table 4.2), the moderation analysis was repeated only for blast
intensity delivered under provoked aggression. For clarity of presentation, the partial
effects of covariates are presented separately (Table 4.5). Under low provocation,
mindfulness continued to moderate the effect of depletion on blast intensity (c3: B =
-1.42, SE = .70, p < .05). The total variance uniquely attributable to the interaction
slightly dropped from 3.79 to 3.13% (F(1, 102) = 4.10, p < .05). With the inclusion of
SP, blast intensity on low provocation was affected by significant main effects of
depletion (c1: B increased from .62 to .73, p < .05), and mindfulness induction (c2: B
slightly dropped from -.62 to -.67, p < .05). Those higher in SP also delivered higher
levels of blast intensity under low provocation.
Under high provocation, there was still no significant interaction effect (c3: B = -.89,
SE = .81, p = .27), main effect of mindfulness (c2: B dropped from .42 to -.74, p = .07),
or depletion (c1: B increased from -.62 to -.67, p = .10) on blast intensity. The
proportion of the total variance in the outcome due to the interaction dropped from
2.07% to 1.00% (F(1, 102) = 1.22). None of the hypothesised covariates significantly
predicted blast intensity under high provocation.
Supporting Hypothesis 3, controlling for personality variables yielded a similar
pattern of results for the moderation of mindfulness induction on the link between
depletion and blast intensity on no provocation, and low provocation, but not on
high provocation. In addition, depleted participants delivered higher blast intensities
except on high provocation.
Table 4.5Partial effects of control variables for moderation models with bootstrapping method, usingdepletion as predictor, mindfulness as moderator, blast intensity as outcome
CovariatesLow provocation High provocation
B SE B SE
Sex -.67 .38 - -AQ Physical aggression .19 .25 .48 .26STAR Provocations .55* .25 .29 .28TMS Curiosity - - .35 .25
133
CovariatesLow provocation High provocation
B SE B SE
PANAS NA2 .68 .49 .57 .55Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
4.4.5.2 Maximum blast latency
During the TCRT, only 77 participants out of 110 delivered maximum blasts to the
opponents. We firstly carried out censored survival analyses using Cox regressions to
assess the extent to which individuals were more likely to administer the maximum
blasts as a function of the experimental conditions. Our proposed model was not
significant (2 = 2.93, p = .40). The likelihood of delivering the maximum blast to
opponent was not predicted by mindfulness (Wald = 1.41, p = .23, CI [.34, 1.31],
exp(B) = .6), depletion (Hypothesis 1b: Wald = .51, p = .47, CI [.42, 1.50], exp(B) =
.79), or the interaction between mindfulness and depletion (Hypothesis 2b: Wald =
2.68, p = .10, CI [.85, 5.86], exp(B) = 2.24).
We repeated the censored survival analyses to assess the maximum blast latency as
a function of the experimental conditions and the hypothesised covariates (i.e., sex
and physical aggression). Our final model was significant (2 = 17.63, p < .01},
supporting Hypothesis 3. The point at which participants delivered the maximum
blast to opponent was influenced by the interaction of mindfulness and depletion
(Wald = 5.32, p < .05, CI [1.20, 8.86], exp(B) = 3.25) and the main effect of sex (Wald
= 8.60, p < .01, CI [1.28, 3.41], exp(B) = 2.09). Interestingly, depleted participants
with mindfulness induction, and females were more likely to administer the
maximum blast during the TCRT. The influence of physical aggression was
approaching significance (Wald = 3.79, p = .052, CI [1.00, 1.89], exp(B) = 1.37), but
not the main effect of depletion (Wald = 2.13, p = .14, CI [.32, 1.18], exp(B) = .61) or
mindfulness (Wald = 2.16, p = .14, CI [.29, 1.19], exp(B) = .60).
When we analysed the effect of mindfulness and ego depletion on maximum blast
latency separately for males and females, the findings varied (Figure 4.4).
134
Figure 4.4 Sex differences in maximum blast latency. Error bars represent standard error ofmean, asterisk indicates that participants delivered the maximum blast sooner (p < .01).
For females (n = 36), the depletion x mindfulness induction model was significant
(2 = 7.95, p < .05). Depleted females who received mindfulness induction were
more likely to deliver the maximum blast (Wald = 5.81, p < .01, CI [1.42, 29.98],
exp(B) = 6.65). There were no significant main effects of depletion (Wald =.41, p =
.52, CI [.28, 1.92], exp(B) = .73), or mindfulness induction (Wald = 3.40, p = .07, CI
[.11, 1.07], exp(B) = .35) on maximum blast latency.
For males (n = 41), the depletion x mindfulness induction model was not significant
(2 = 1.19, p = .76). Maximum blasts latency was not predicted by depletion (Wald =
.58, p = .45, CI [.30 to 1.71], exp(B) =.71), mindfulness induction (Wald = .02, p = .89,
CI [.43, 2.64], exp(B) = 1.06), or interaction between conditions (Wald = .24, p = .62,
CI [.37, 5.31], exp(B) = 1.40).
4.4.5.3 Correlation between blast intensity and maximum blast latency
For participants who had delivered the maximum blast (N = 77), there were negative
correlations between maximum blast latency and blast intensity across levels of
provocation (r = -.43 on no provocation, r = -.59 on low provocation, and r = -.39 on
high provocation; all significant at p < .0001). Specifically, higher blast intensities
were delivered by those who had delivered the maximum blast to the opponent
135
earlier in the TCRT. The correlation persisted when analysed separately for males
and females, except on high provocation. On high provocation, the correlation
between maximum blast latency and blast intensity became non-significant for
females (r = -.30, p = .08).
4.4.6 Moderation of mindfulness induction on the link between depletion and
indirect aggression
The moderating effect of mindfulness induction on the link between depletion and
each indirect aggression item was analysed separately, also using a bootstrapping
method (Table 4.6). As in Study 2, to survive Bonferroni corrections (p < .01), we
used 99% bias-corrected confidence intervals. Mindfulness induction did not
moderate the relationship between depletion and indirect aggression (Hypothesis
2c). Additionally, there were no significant main effects of depletion or mindfulness
induction on indirect aggression (Hypothesis 1c).
Table 4.6.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, rating of opponent as outcome
Model testedAggressive Skilful Competitive Fair Play again
B SE B SE B SE B SE B SE
Depletion (c1) .20 .24 -.09 .24 .09 .24 .40 .32 -.02 .30Mindfulness induction(c2)
-.06 .24 .06 .23 .16 .24 -.04 .32 .24 .30
Depletion x mindfulness(c3)
.03 .48 -.25 .47 -.04 .47 -.11 .63 -.72 .60
R2 on interaction 0% 26% 0% .30% 1.31%Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
As suggested in section 4.4.3, no sex differences were found in any of the indirect
aggression ratings (p = .06 to 50), so we did not break down moderation models by
sex (see Table 4.7 for sex differences in ratings of indirect aggression).
136
Table 4.7.Sex differences based on rating of opponent
Indirect aggressionMales (n = 52) Females (n = 58)
tM (SD) M (SD)
Aggressive 3.94 (1.31) 4.24 (1.16) t(108) = -1.28, p = .21Skilful 4.75 (1.37) 5.03 (1.01) t(108) = -1.22, p = .23Competitive 4.87 (1.31) 5.31 (1.10) t(108) = -1.94, p = .06Fair 4.40 (1.87) 4.71 (1.41) t(94.46) = -.95, p = .35Play again 5.46 (1.70) 5.26 (1.46) t(108) = .67, p = .50*p < .05;
**p < .01;
***p < .001
The moderation analyses for ratings of the opponent skill, fairness, and participants
willingness to play against the same opponent were repeated by including the
hypothesised covariates based on the prior significant zero-order correlations
between covariates and rating indirect aggression (Table 4.8; see again Table 4.2, in
which no personality variables were significantly associated with ratings of opponent
aggressiveness and competitiveness). Nevertheless, there was no effect of
experimental conditions on ratings for these items (Hypothesis 3). Those who were
higher in NA following depletion task rated the opponent as being less skilful, and a
less willingness to take part in a task with the same opponent again. The positive
impact of curiosity on participants’ willingness to take part in a task with the same
opponent again nearly survived Bonferroni correction (p < .011).
Table 4.8.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, rating of opponent as outcome, controlling for covariates
Model testedSkilful Fair Play again
B SE B SE B SE
Depletion (c1) -.01 .23 .52 .31 .13 .28Mindfulness induction (c2) .01 .23 -.13 .31 .06 .28Depletion x mindfulness (c3) -.40 .47 -.21 .62 -.40 .58R2 on interaction .65% .09% .39%Covariates:MAAS - - -.23 .27 - -SCS - - -.41 .34 - -AQ Total .42 .31 - - - -AQ Verbal aggression .10 .21 - - - -STAR Provocations - - .35 .32 - -STAR Frustrations - - .03 .32 - -
137
Model testedSkilful Fair Play again
B SE B SE B SE
TMS Curiosity - - - - .45* .17PANAS NA1 -1.09** .37 - - -1.70** .45Self-harm - - .62 .52 - -Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
4.4.7 Correlation between blast intensity and indirect aggression
A significant correlation between blast intensity and indirect aggression was found in
rating of the opponent’s fairness, such that participants evaluate the opponent that
they had blasted with more intense blasts as being more fair (Table 4.9).
Table 4.9.
Bivariate correlations between blast intensity and rating of opponent
Indirect aggression No provocation Low provocation High provocation
Aggressive -.05 -.05 .00
Skilful .01 .06 .14
Competitive -.09 -.15 -.12
Fair .08 .24* .27**
Play again -.02 .05 .10*p < .05;
**p < .01;
***p < .001
Table 4.10.
Correlations between blast intensity and rating of opponent based on sex
*p < .05;
**p < .01;
***p < .001
138
When we broke down the correlations by sex (Table 4.10), we found that the
correlation between blast intensity and rating of opponents’ fairness was more
pronounced in males than in females. Females evaluated opponents that they had
blasted with more intense blasts as being less competitive.
4.4.8 Moderation of mindfulness induction on the link between depletion and
self-control performance
Table 4.11.Overall change in handgrip performance
Depletion No depletion
Change 1 Change 2 Change 1 Change 2
Mindfulness induction -7.04 -6.25 -2.33 -2.56No mindfulness -17.93 -12.41 -9.86 -12.61Note. Higher positive scores indicate better self-control on all measures. Values represent thechanges in mean time that participants squeezed the handgrip pre-TCRT task (Change 1) and post-TCRT task (Change 2).
Table 4.11 shows the changes in self-control performance as measured by mean
differences between handgrip duration at baseline and pre-TCRT task (Change 1) and
post-TCRT task (Change 2). Change 1 and Change 2 were positively correlated (r =
.72, p < .0001).
Table 4.12.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, changes in handgrip performance as outcome, controlling for covariates
Model testedChange 1 Change 2
B SE B SE
Depletion (c1) -5.14 3.33 -3.20 3.22Mindfulness induction (c2) 9.23** 3.27 8.26** 3.17Depletion x mindfulness (c3) .69 6.56 -4.68 6.34R2 on interaction .01% .44%Covariates:Sex 13.98*** 3.28 6.22 3.33SCS -5.82 3.03 - -AQ Physical Aggression - - -5.23* 2.05Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
Change 1 = mean time differences between handgrip duration at baseline and pre-TCRT task; Change2 = mean time differences between handgrip duration at baseline and post-TCRT task*p < .05;
**p < .01;
***p < .001
139
The significantly correlated personality factors (i.e., sex and trait self-control at
Change 1; sex and trait physical aggression at Change 2; see again Table 4.2) were
entered into the moderation models (Table 4.2). The effect of depletion at Change 1
(Hypothesis 4a) and Change 2 (Hypothesis 4b) on handgrip performance was not
moderated by mindfulness induction. Mindfulness induction alone predicted
handgrip performance at both time points after controlling for these covariates. In
addition, a better handgrip performance at Change 1 was shown by females. A
better handgrip performance at Change 2 was shown by those with lower levels of
physical aggression. Sex differences in handgrip stamina are presented in Figure 4.5.
Figure 4.5. Sex differences in handgrip stamina. Higher positive scores indicate better self-control in Change 1 (pre-TCRT task) and Change 2 (post-TCRT task). Error bars represent
standard error of mean, asterisk indicates significant (p < .001) sex differences in handgripstamina.
4.5 DISCUSSION
4.5.1 Effect of mindfulness induction on aggressive behaviour following depletion
The current study employed manipulations of ego-depletion and mindfulness, using
110 (58 females) native British students. As expected, the first main analysis
revealed that following no provocation and low provocation trials, depleted
140
participants who received mindfulness induction delivered lower levels of blast
intensity to the opponent compared to depleted ones with no mindfulness
induction. This finding supports our proposal that the negative effect of depletion on
people’s tendency to behave aggressively might be counteracted by mindfulness
interventions. Additionally, under high provocation trials, the moderating effect of
mindfulness on the depletion and direct aggression link diminished, and so did the
main effect of depletion on direct aggression. This latter finding is in line with past
research showing that a very intense level of provocation is commonly sufficient to
decrease the effect of a number of variables, such as sex differences (Bettencourt &
Miller, 1996), and initially non-aggressive opponent’s behaviour (Lawrence &
Hutchinson, 2013a, 2013b) on direct aggression (i.e., ceiling effect). Our next
experiment will examine if a similar effect also occurs in a non-Western sample.
The second main analysis revealed a reverse effect of mindfulness induction on
maximum blast latency following depletion in females. Specifically, the likelihood of
delivering the maximum blast earlier increased for depleted females with
mindfulness induction as opposed to depleted females with no mindfulness
induction—although this tendency was not generally accompanied by higher blast
intensity. The decrease in maximum blast latency may, be explained in two ways.
Depleted females who received mindfulness induction may have not only fully
attended to the experience of being provoked, but also responded in a reactive way
that was inconsistent with mindfulness. For novice practitioners, the ability to
observe internal and external experiences may be associated with nonsignificant and
sometimes even positive correlations with psychological symptoms, suggesting
inclination to observe in a judgmental way (Baer et al., 2008). Future studies could
include a measure of mindfulness on the facets of observing and nonjudging (e.g.,
Baer et al.’s Five Facet Mindfulness Questionnaire [FFMQ], 2006), to test the
possible sex differences on these facets.
141
Sex differences in maximum blast latency might also be related to the subtle/implicit
nature of the latency measure. For instance, previous TCRT work looking at shock
intensity and shock duration found that under the influence of alcohol, females
delivered only higher levels of shock duration, whereas their counterpart males
expressed aggression in both measures (Giancola & Zeichner, 1995). Evers, Fischer,
Mosquera, and Manstead (2005) also found that females expressed lesser anger
(i.e., delivered less hot sauce to a partner) than males when they expected negative
social consequences, but no sex differences were found in the nonsocial condition
(see also Eagly & Wood, 2009). Delivering earlier maximum blast might be a way for
females of being aggressive, but less obviously so, if they feel less judged.
The prior literature suggests that the sex differences in aggression usually narrows
when indirect aggression is measured. Accordingly, significant sex differences were
found in terms of blast intensity on low provocation, and approaching significant on
blast intensity on no provocation and on maximum blast latency. No sex differences
were found on ratings of indirect aggression. Moreover, ratings of indirect
aggression for both males and females were not predicted by any of the
experimental conditions, but were related to negative affect following depletion.
Those who were higher in negative affect rated the opponent as being less skilful
and expressed less willingness to take part in a task with the same opponent again.
This is quite a common finding, since aggressive behaviour is typically triggered by
negative affect (e.g., anger) produced by unpleasant experiences (Anderson &
Bushman, 2002; Berkowitz, 1989, 1990, 1993).
In addition, both traits of mindfulness and self-control were insubstantially
associated to aggressive behaviour, although they were negatively correlated with
individual propensity to aggress (i.e., trait aggression, sensitivity to provocations [SP]
and frustrations [SF]; similar to the finding in Study 1 and 2). Self-harm was even not
correlated to any of self-reported measures or to aggressive behaviour. Thus the role
of personality variables might even be less salient when aggressive behaviour were
measured following depletion. However, individual differences in SP did predict
142
higher levels of blast intensity under low provocation (similar to Lawrence &
Hutchinson, 2013b). The significant link between depletion and blast intensity under
low provocation was also activated after the inclusion of SP. This finding (also in
Study 2) generally supports Lawrence’s (2006) notion that SP may generate specific
predictions about who is likely to behave aggressively to provoking triggers.
4.5.2 Effect of mindfulness induction on self-control performance following
depletion
There was a main effect of mindfulness on self-control performance (i.e., handgrip
stamina/changes in handgrip duration relative to baseline) when self-control
performance was measured before the TCRT task. Specifically, participants who
performed mindfulness exercise outperformed those who did not perform this
exercise. This effect, however, was independent of the effect of depletion condition.
In fact, no usual effect of depletion condition was found on self-control
performance. Females also outperformed males.
The main effect of mindfulness induction persisted when self-control performance
was measured again after the TCRT task. At this point, there was still no moderating
effect of mindfulness induction on the link between ego-depletion and self-control
performance. Here again, depletion condition did not influence participant’s ability
to perform self-control task. Altogether, it seems in the current sample, the effect of
mindfulness induction on self-control performance was less predictable by the
mechanism of self-control.
4.5.3 Strengths and limitations of the current study
Recently, Friese et al. (2012) have demonstrated the benefit of mindfulness
induction on self-control ego-depletion outside the area of aggression, but did not
explore the effect for non-depleted individuals. The current study extended their
143
work via a fully crossed design, establishing that in the absence of depletion, the
effect of mindfulness induction on aggressive behaviour and self-control
performance would be less salient. As in Study 2, three indicators of aggressive
behaviour (i.e., blast intensity, maximum blast latency, anonymous reputation
damage of opponent) were used in the current study.
The 15-min mindfulness breathing exercise used in the current study was successful
in increasing mindfulness state of decentering, but not state of curiosity. According
to Lau et al. (2006), curiosity is typically associated with mindfulness meditation
experience, in which those with more than 1 year of experience may likely score
significantly higher on TMS curiosity than those with less than 1 year of experiences.
Therefore, the non-significant changes in curiosity might be related to the current
participants’ lack of familiarly with mindfulness procedures (only 1 participant
reported practising mindfulness on monthly basis). On the other hand, decentering is
close to Shapiro et al.’s (2006) conceptualisation of reperceiving (i.e., a shift in
perception that the experiences perceived by the individuals are separate from the
individuals themselves). As discussed in Chapter 1, reperceiving enables individuals
to become more tolerant to unpleasant experiences related to depletion and
aggression-related emotions.
As in Study 2, an important limitation of the current study is that we failed to recruit
the optimum number of 128 participants (as calculated by G*Power 3, 2013), albeit
our 110 final sample was sufficient with reference to the previous similar study
(Friese et al., 2012). Moreover, unlike in Study 2, control participants were not given
relaxation procedure, but were asked to listen to a neutral educational information
and to complete a word search puzzle (e.g., Erisman & Roemer, 2010). Similar to the
control condition for non-mindfulness induction participants in Friese et al.’s (2012)
ego-depletion study, these types of task still require participants to pay attention
and concentrate, but without inducing additional depleting effect.
144
4.5.4 Conclusions and next step
Findings from the current study suggested that when people’s self-control resource
has been depleted, mindfulness induction may reduce their aggressive behaviour in
terms of blast intensity delivered to the opponent under no provocation and low
provocation trials. Individual differences in SP also predicted provoked aggression.
Further, sex differences emerged on blast intensity, maximum blast latency, and self-
control performance, but not on ratings of indirect aggression. The same
experimental study using cross-cultural sample, specifically Indonesian university
students, will be presented in the following chapter. Mindfulness is derived from
Buddhist and Eastern philosophies and practices. Therefore, it is important to
examine the effects of mindfulness interventions following depletion on aggression
beyond the Western sample. It may be that the effects of mindfulness seen so far in
the thesis are the result of the novelty or salience of an Eastern practice being used
in a Western environment and culture. The next chapter asks: does mindfulness on
aggression replicate in a non-Western culture. In this way it is also possible to
examine whether the insignificant effect of experimental conditions on indirect
aggression, as well as the relatively trivial role of personality variables on aggressive
behaviour would differ across cultures.
145
CHAPTER FIVE
Study 4: Effect of mindfulness induction on aggressive behaviour following
depletion (Indonesian sample)
5.1 INTRODUCTION
In the previous study using British sample (Chapter 4), depleted participants who
received mindfulness induction displayed less direct aggression (as measured by
levels of blast intensity under no provocation and low provocation in the adapted
Taylor Competitive Reaction Time [TCRT]: Taylor, 1967, task) as opposed to depleted
participants who had not received mindfulness induction. The moderating effect of
mindfulness on indirect aggression and self-control performance following depletion
was less observable, and was even reversed on the maximum blast latency for British
females. The purpose of the present study is to (i) replicate the findings in Chapter 4,
and (ii) examine if the mindful effect on depleted participants holds in a sample
taken from an Eastern tradition and culture.
5.1.1 Mindfulness and Indonesian culture
Mindfulness practices originate in the Eastern contemplative traditions, Buddhism in
particular. However, even cross-cultural validation of mindfulness interventions
derived from Western operationalisation is currently lacking (Christopher et al.,
2009). As a result, this current study replicates the previous study examining the
impact of ego depletion and mindfulness intervention on aggressive and self-control
behaviour in Indonesian groups.
The application of mindfulness in Indonesia may be reflected in the Javanese’s (the
largest ethnic groups in Indonesia) endorsement of the “eling” value
(Koentjaraningrat, 1989), which means continuously being aware of one’s own
position in life. Javanese children are conventionally educated to develop a
perception of having “enough” (as opposed to wanting more), so that they can
146
always be aware of all the difficulties and failures they may face in life rather than be
driven by the pursuit of goals (Koentjaraningrat, 1989). As noted by Brown et al.
(2007a), while mindfulness is not antithetical to the quest of self-controlled goals,
the mindful goals are more selective in accord with one’s intrinsic values.
5.1.2 Indonesian culture and aggression
In the aggression literatures, one pertinent variable that influences cultural
differences in their expression and tolerance of conflict, aggression, and anger is
individualism vs. collectivism values (Hofstede, 1980; Triandis, 1995). Individualism
emphasises the independence from groups and a greater sense of obligation to the
individual, whereas collectivism imposes the interdependence of individuals and a
greater sense of obligation to the group. As a result, aggression has been
conceptualised as a means to win competitions and to achieve self-reliance within
the individualistic, but as a disruptive act within the collectivistic cultures (Li et al.,
2010).
Hofstede and colleagues (Hofstede, 2010; Hofstede, Hofstede, & Minkov, 2010) have
developed a map of individualism, based on replications and extensions of a
previous study using IBM employees from 76 countries and study using World Value
Survey from representative samples of national populations from 93
countries, where high scores indicate more individualistic cultures/countries and
lower scores more collective cultures/countries. This map yields a score of 14 for
Indonesia vs. 89 for United Kingdom. The Indonesian Javanese culture strongly
prohibits rude conduct, shouting, or open conflict since it is presumed as a sign of
lack of self-control and inner strength; and conversely highly values the ability to
speak unpleasant subjects in an indirect manner to preserve the impression of
harmonised social relationships (Koentjaraningrat, 1985; Magnis-Suseno, 1997).
These features are also evident in the Sundanese population (see French, Jansen &
Pidada, 2002), the second largest ethnic group in Indonesia. Theoretically, this may
reduce the Indonesians’ display of direct aggression, particularly physical aggression,
compared with societies with stronger individualistic values such as the UK. Past
147
cross-cultural review (Bergeron & Schneider, 2005) has revealed that peer-directed
aggression was positively related to Hofstede’s individualism dimension.
As a consequence, however, it is plausible that Indonesian people also have lower
tolerance to indirect aggression. Forbes et al.’s (2009) study comparing college
students in the US (individualist), Poland (mid scoring), and China (collectivist) found
that for both direct and indirect aggression, increases in individualism were
associated with increases in aggression. Relational or indirect aggression may be
unacceptable since it also threatens interpersonal relationships and group harmony
(Li et al., 2010). At the same time, the opposite prediction is possible. The need to
display harmony may not correspond to the actual individual’s emotional states, and
could be articulated in the adoption of more indirect methods of retaliation for the
collectivist (French et al., 2002). As relational or indirect aggression is more covert,
the expression of relational aggression is less seen and therefore not so open to
negative evaluations from others.
5.1.3 Indonesian culture and self-control
The attributes associated with individualism and collectivism may have
straightforward implications for self-control. As discussed in Chapter 1 (section
1.2.3.1), self-control strength is only one of the four major components of self-
control (for a full review, see Baumeister, Schmeichel, et al., 2007 and Baumeister &
Vohs, 2007). The other three are standard, monitoring, and motivation. Each of
these components is necessary for effective self-control, and can compensate for
each other to some extent. Thus theoretically, depleted individuals should still be
able to exert control provided they have sufficient motivation to conform either to
societal expectations or to their own personal values.
Seeley and Gardner (2003) suggested that because individuals high in social
orientation (e.g., collectivist cultural background and interdependent beliefs) would
be more motivated to engage in self-control in daily social interaction, they may
develop greater ability for self-control and become less prone to ego-depletion.
148
Accordingly, they demonstrated that a thought suppression task reduced handgrip
performance amongst native-born US students (i.e., individualistic cultural
background) but had no effect on handgrip performance amongst foreign-born Asian
students (i.e., collectivism cultural background). A more recent study also showed
that social value orientation moderates the impact of depletion (i.e., suppressing
emotion while watching a comedy video) on concern with others’ well-being in
terms of points allocated to others in an interdependent decision-making task
(Balliet & Joireman, 2010). These findings suggest that the presence of collectivism
value in the Indonesian people may increase their ability to exert control and their
resistance to depletion.
It should be noted, however, that self-control in terms of regulating one’s behaviours
to meet their personal goals and professional achievement is pervasive in Western
cultures (Brown et al., 2007a). Thus it is also plausible for the Indonesians to be less
familiar to the conceptualisation of self-control derived from Western cultures.
5.1.4 Role of mindfulness induction in counteracting the effect of depletion on
aggressive behaviour for Indonesian culture
Altogether, a greater familiarity with mindfulness in daily life, lower tolerance to
aggression display, and higher resistance to depletion may strengthen the effect of
mindfulness induction for the Indonesian. Hence, the moderating role of
mindfulness found in Chapter 5 (under conditions of ego-depletion) should also be
also seen in the Indonesian sample, in terms of direct aggression (following low
provocation in particular) and self-control performance (handgrip stamina). Because
the sample in Indonesia is accustomed to mindful ways of thinking, however, it is
also plausible that no additional impact of the mindful intervention would be seen.
Additionally, the application of the Taylor paradigm (TCRT task) may be less useful in
the Indonesian sample, as the sample has strict sanctions against the use of violence
or aggression. To our knowledge, the Taylor paradigm has not been used in
Indonesian samples to date, and so this is the first study to examine the
149
effectiveness of the paradigm here. It is expected that the level of blast intensity
should increase with additional provocation in order to demonstrate some validity
for the paradigm. Also, as males and females typically differ in aggressive behaviour
(e.g., Archer, 2004; Bettencourt & Miller, 1996; Giancola & Parrot, 2008), we should
expect to see males delivering higher intensity blasts under conditions of no and low
provocation during the TCRT in Indonesia as in the British sample.
5.2 HYPOTHESES
Similar to Study 3 (see section 4.2), our general hypothesis is that that ego-depletion
will predict higher levels of aggression and impaired self-control performance
amongst participants with no mindfulness induction but not amongst participants
with mindfulness induction, after accounting for the potential influence of individual
differences in mindfulness, self-control, aggression (i.e., trait aggression, sensitivity
to provocations [SP] and frustrations [SF]), sex, and self-harm.
5.3 METHODS
5.3.1 Participants
A total of 124 students from University of Brawijaya, Indonesia entered the current
study. Based on sex, participants were randomly assigned to one of the four
experimental groups (2: mindfulness [intervention vs. no intervention] x ego-
depletion [depletion vs. no depletion]). Five participants were discarded from final
analysis due to: expressing spontaneous suspicions expressing spontaneous
suspicions regarding the TCRT task (3 participants) and falling asleep during the
mindfulness breathing task (2 participants), resulting in 119 participants (60
females). Participants’ mean age was 20.40 (SD = 1.24), 86.55% belong to Javanese
ethnic group, 3.36% Sundanese, and the remainder were of unspecified Indonesian
ethnicity. None of the participants reported having current encounter with
mindfulness experiences. The current study lasted from February until March 2012.
150
5.3.2 Design
Figure 5.1. Flow of participants through Study 4.Note. M = Males; F = Females
As in the previous study, a 2 (ego-depletion vs. no depletion) x 2 (mindfulness
induction vs. no induction) experimental design, along with stratified sampling
technique was employed again in this study. Figure 5.1 depicts flow of participants.
5.3.3 Trait and state measures
The translation of the trait and state measures in the current study was made by two
postgraduate Indonesian students, and was back-translated into English and checked
against the original transcript by a commercial translation service (Viesta translator).
The adapted measures were then pre-tested to 2 pilot participants. We used the
151
same set of trait measures from Study 1, 2, and 3 (Appendix 2.1), the state measures
from Study 2 and 3 (Appendix 3.1), and the two questions on the efficacy of the
depletion manipulation from Study 3 (i.e., task difficulty and the amount of attention
control). The description of these measures is provided in section 2.3.3 and 3.3.3.
Following the development of this thesis, cross-validation for some of the
aforementioned trait measures is being provided in Asian countries. Specifically, the
MAAS (Brown & Ryan, 2003) was recently shown as a sound measure of trait
mindfulness in Chinese adolescents (Black, Sussman, Johnson, & Milam, 2012), the
AQ (Buss & Perry, 1992) was used amongst female prisoners in Malay (Mazlan &
Ahmad, 2012), and the DSHI (Gratz, 2001) was used in university students in
Indonesia (Tresno et al., 2012). This supports the idea that Western
operationalisations of the concepts of mindfulness and aggression may be related to
those concepts in non-Western settings.
5.3.4 Experimental manipulations
The current study employed the same ego-depletion task, mindfulness induction, the
adapted Taylor Competitive Reaction Time (TCRT: Taylor, 1967) task, and self-control
performance/handgrip task from Study 3 (see section 4.3.4).
5.3.5 Power calculation
As in Study 3, this study required a total of between 88 (according to the Friese et
al.’s study in 2012) to 128 participants (according G*Power 3, 2013, calculation for a
medium effect [d = .25], given the power of .80 and an alpha level of .05).
5.3.6 Procedure
Potential participants from University of Brawijaya, East Java, Indonesia were
contacted via posters/leaflets on campus. As in Study 3, participants were told that
152
the study aimed to investigate the effects of personality traits and experimental
treatments on the way people perform in a competitive reaction-time task
(Appendix 4.2). They were recruited in exchange for inconvenient allowance (equals
to £1) and a chance to win an incentive for the fastest participants’ reaction time
across the entire study (equals to £10). These amounts are equivalent to the British
participants’ incentive in terms of real value.
Participants came individually into the Psychology lab at University of Brawijaya,
Indonesia. The rest of the procedure followed that outlined in Study 3 (see section
4.3.6, Chapter 4).
5.3.7 Data analysis
Statistical analyses (as described in Study 3, section 4.3.6, Chapter 4) were
performed using IBM SPSS statistics 20.0 (2011). Importantly, the moderating effect
of mindfulness induction on the link between self-control ego depletion and
aggressive behaviour (primary hypotheses) and between depletion and self-control
performance (secondary hypotheses) were tested with bootstrapping method, using
model number 1 in the A. F. Hayes’s (2012a) PROCESS macro for simple moderation.
5.4 RESULTS
5.4.1 Baseline levels
The four experimental groups (i.e., depletion plus mindfulness, depletion no
mindfulness, no depletion mindfulness, no depletion no mindfulness) were not
significantly different from each other in means of the trait measures (p = .08 to .86),
indicating groups equivalency at baseline levels.
None of the participants reported having had experience with mindfulness practice,
thus further comparisons based on this factor could not be carried out.
153
5.4.2 Preliminary analysis
Table 5.1 presents the zero-order correlations amongst trait measures and
descriptive statistics (N = 119). Internal reliabilities ranged from adequate ( = .69)
to high ( = .88), except for verbal aggression ( =.50). Deleting any of the verbal
aggression items did not increase its reliability ( = .38 to .52). We decided to retain
this subscale in the calculation of a total or general trait aggression score since it was
significantly related to two out of three other subscales in the AQ as in the majority
of previous studies using this scale (i.e., physical aggression and anger). Moreover,
excluding this subscale did not alter the reliability for the total aggression ( only
increased to .82 from .81).
Importantly, as in Study 1, 2, and 3, trait mindfulness and trait self-control were
positively correlated to each other. Self-control was negatively correlated to all of
the self-reported aggression except verbal aggression, whereas mindfulness was only
negatively correlated to anger and hostility.
Table 5.1.Zero order correlations and psychometric properties of measures
Measures 1 2 3 4 5 6 7 8 9
MAAS (1) 1.00SCS (2) .48** 1.00Total AQ (3) -.28** -.47*** 1.00AQ Physical (4) -.15 -.36*** .72*** 1.00AQ Verbal (5) -.04 -.05 .55*** .30** 1.00AQ Anger (6) -.24** -.47*** .83*** .47*** .26** 1.00AQ Hostility (7) -.34*** -.38*** .62*** .24** -.02 .46*** 1.00STAR Provocations (8) -.02 -.36*** .52*** .33*** .18 .47*** .42*** 1.00STAR Frustrations (9) -.07 -.39*** .49*** .23* .15 .51*** .41*** .83*** 1.00
M 3.93 3.14 2.76 2.92 2.77 2.14 3.14 2.67 3.09SD .57 .46 .45 .78 .82 .61 . 63 .76 .64
Cronbach .77 .74 .81 .87 .88 .69 .50 .71 .72
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = AggressionQuestionnaire; STAR = Situational Triggers of Aggressive Responses*p < .05;
**p < .01;
***p < .001
Compared to females (n = 60), males (n = 59) scored significantly higher on physical
aggression (t(104.83) = 3.34, p < .001; M males = 2.32, SD = .67 vs. M females = 1.96,
SD = .48) and lower on sensitivity to frustrations (SF: t(117) = -2.51, p < .05; M males
154
= 2.58, SD = .78 vs. M females = 2.95, SD = .83). Compared to non self-harmers (n =
97), self-harmers (n = 22) scored higher on general trait aggression (t(117) = 2.00, p <
.05; M self-harmers = 2.93, SD = .47 vs. M non self-harmers = 2.72, SD = .44), physical
aggression (t(117) = 2.03, p < .05; M self-harmers = 2.37, SD = .68 vs. M non self-
harmers = 2.09, SD = .58), and hostility (t(117) = 2.07, p < .05; M self-harmers = 3.34,
SD = .69 vs. M non self-harmers = 3.03, SD = .62), as well as on sensitivity to
provocations (SP: t(117) = 3.22, p < .001; M self-harmers = 3.38, SD = .86 vs. M non
self-harmers = 2.80, SD = .73) and frustrations (SF: t(117) = 2.51, p < .05; M self-
harmers = 3.15, SD = .91 vs. M non self-harmers = 2.67, SD = .78). However, none of
the associations were significantly altered when we carried out partial correlations
controlling for sex (r = -.05 to .84) or self-harm status (r = .02 to .83).
5.4.3 Correlation between personality variables and experimental outcomes
Similar to Study 3, the experimental outcomes in the current study consisted of (i)
direct aggression: blast intensity on each provocation level and maximum blast
latency, (ii) indirect aggression, and (iii) self-control performance/handgrip stamina
in Change 1 (changes in pre-TCRT task handgrip duration relative to baseline) and
Change 2 (changes in post-TCRT task handgrip duration relative to baseline).
Associations between outcomes and personality variables are shown in Table 5.2.
5.4.3.1 Associations with direct aggression
Positive affect (PA) assessed following depletion condition and following the TCRT
task was significantly related to blast intensity under high provocation. Additionally,
compared to females, males delivered higher blast intensity under no provocation
and low provocation, but not on high provocation (see Table 5.3). No differences
were found between self-harmers and non self-harmers on any direct aggression
behaviour (p = .55 to .84).
155
None of the personality variables were significantly correlated to maximum blast
latency. Maximum blast latency was only associated with sex (t(82.45) = 2.46, p <
.01), such that females (n = 45) waited significantly longer than males (n = 53) before
delivering the maximum blasts (M males = 7.11, SD = 10.08 vs. M females = 12.96,
SD = 12.95). No differences were found between self-harmers (n = 19) and non self-
harmers (n = 79; t(96) = .80, p = .43) on maximum blast latency.
5.4.3.2 Associations with indirect aggression
The rating of the opponent’s aggressiveness was negatively related to physical and
verbal trait aggression and positively related to PA1 (following depletion task) and
PA2 (following the TCRT task). PA at both time points was also positively related to
participants’ willingness to play against the same opponent again; this rating was
also positively related to trait hostility. The rating of opponent’s skilfulness was
negatively related to physical aggression and SP and positively correlated to NA
following depletion. No sex differences were found in any of the ratings (p =.11 to
61). Likewise, no significant differences between self-harmers and non self-harmers
in the ratings (p = .17 to .91)
5.4.3.3 Associations with self-control performance
Self-control performance (i.e., handgrip stamina) at Change 1 was associated with
state mindfulness. Additionally, females outperformed males in Change 2 (t(117) =
2.62, p < .01; M male = -12.80, SD = 17.19 vs. M female = -5.15, SD = 14.61), but not
in Change 1 (t(117) = 1.42, p = .16; M male = -3.68, SD = 19.23 vs. M female = 1.07,
SD = 16.68). No significant differences between self-harmers and non self-harmers in
Change 1 (t(117) = .79, p = .43) and Change 2 (t(117) = .18, p = .86).
156
Table 5.2.Zero order correlations of personality measures, aggressive behaviour, and self-control performance
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression Questionnaire; STAR = Situational Triggers of Aggressive Responses; TMS =Toronto Mindfulness Scale, PANAS 1 = Positive Affect Negative Affect following depletion; PANAS 2 = Positive Affect Negative Affect post-TCRT taskAll correlations were based on N = 119, except for maximum blast latency. The correlations for maximum blast latency was made only for participants who delivered amaximum blast (N = 98).*p < .05;
**p < .01;
***p < .001
157
5.4.4 Manipulation checks
The depletion (attention control) manipulation was effective. Participants in the
attention control condition (n = 59) rated the task as more difficult than those in the
no control condition (n = 60; t(117) = 4.35, p < .0001; M attention control = 3.68, SD
= 1.73 vs. M no control = 2.37, SD = 1.56) and having controlled their attention to a
greater extent than no control participants (t(117) = 4.65, p < .0001; M attention
control = 3.78, SD = 1.77 vs. M no control = 2.33, SD = 1.62). The two questions
measuring depletion condition were positively correlated (r = .46, p <.0001).
The mindfulness breathing task succeeded in increasing participants decentering
(t(117) = 2.36, p < .05; M mindfulness induction = 2.38, SD = .47 vs. M no induction =
2.18, SD =. 47), but not their curiosity (t(117) = 1.65, p = .10; M mindfulness
induction = 2.77, SD = .54 vs. M no induction = 2.61, SD = . 53). Internal consistency
in both scales were adequate for curiosity ( = .76) and fair for decentering ( = .62).
Both scales were positively correlated (r = .61, p < .0001).
Mood after depletion did not differ between depleted and non-depleted participants
(PA1: t(101.65) = -.75, p = .45; NA1: t(117) = .20, p = .85). Both scales showed good
internal consistency ( = .79 and .81 for PA1 and NA1, respectively), but were not
correlated (r = -.02, p = .80). Similarly, mood following the TCRT did not differ
between depleted and non-depleted participants (PA2: F(3,115) = 1.57, p = .20; NA2:
F(3,115) = 1.05, p = .37). Both scales showed good internal consistency ( = .84 for
both PA2 and NA2) and were positively correlated (r = .28, p < .005). Results from
both mood measures indicated that the experimental outcomes would not be
affected by fluctuations in participants’ mood following the manipulations.
158
5.4.5 Moderation of mindfulness induction on the link between depletion and
direct aggression
5.4.5.1 Levels of blast intensity
Figure 5.2. Effect of level of provocations on blast intensity.Error bars indicate standard error of mean. Scale goes up to 8 as a maximum.
A one-way repeated-measure ANOVA was performed to test differences in blast
intensity under no provocation, low provocation, and high provocation trials (Figure
5.2). Unusually, blast intensity was not affected by level of provocation (F(1.56,
183.69) = 1.88, p = .17). Participants did not deliver higher blast intensity under low
provocation (M = 5.32, SD = 1.78) compared to under no provocation (M = 5.61, SD =
2.20; p = .06), or under high provocation (M = 5.37, SD = 1.94) compared to under
low provocation (p =.63).
A comparable pattern of results was shown when we included sex (a 2 [sex] by 3
[provocation level] mixed analysis ANOVA), in which blast intensity was neither
affected by level of provocations, (F(1.55, 181.64) = 1.91, p = .16) nor by interaction
between sex and level of provocations (F(1.55, 135.95) = 2.04, p = .14). It should be
noted, however, that the typical sex differences in the TCRT task (see Giancola &
Parrott, 2008) occurred in this sample, in that males delivered higher intensity blasts
under conditions of no and low provocation but not under high provocation
compared to females (Table 5.3).
159
Table 5.3.Sex differences on blast intensity on each provocation level
Level of provocationsMales (n = 59) Females (n = 60)
tM (SD) M (SD)
No provocation 6.03 (2.05) 5.20 (2.30) t(117) = 2.09*Low provocation 5.70 (1.83) 4.94 (1.67) t(117) = 2.36*High provocation 5.49 (2.17) 5.26 (1.68) t(109.28) = .65, p = .52*p < .05;
**p < .01;
***p < .001
As in Study 3, the moderation model of mindfulness on the link between depletion
and blast intensity (Table 5.4) was tested using a bootstrapping method (model
“number 1” in A. F. Hayes’s (2012a) PROCESS macro), with 95% bias-corrected
confidence intervals (based on 5,000 bootstrap resamples, N = 119). All predictors
were mean-centered. Mindfulness induction significantly moderated the link
between depletion and blast intensity under low provocation only (Hypothesis 2a).
Specifically, following low provocation, depleted participants with mindfulness
induction delivered lower levels of blast intensity compared to depleted ones
without induction, whereas non-depleted participants with mindfulness induction
delivered comparable blast intensity with non-depleted ones without induction (See
Figure 5.3). Additionally, there was a significant main effect of depletion on blast
intensity under low provocation (Hypothesis 1a).
Table 5.4.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, blast intensity as outcome
Model testedNo
provocationLow
provocationHigh
provocation
B SE B SE B SE
Depletion (c1) .30 .40 .63 .32* .52 .35Mindfulness induction (c2) -.50 .40 -.35 .32 -.33 .35Depletion x mindfulness (c3) -1.28 .80 -1.28 .63* -1.23 .70R2 on interaction 2.11% 3.25%* 2.52%Conditional effect of depletion withoutmindfulness
1.29** .45**
Conditional effect of depletion withmindfulness
-.001 .45
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
160
Figure 5.3. Mindfulness × depletion effect on blast intensity across provocation levels.Error bars indicate standard error of mean, asterisk indicates significant (p < .05) decrease in
blast intensity. Scale goes up to 8 as a maximum.
The moderation analysis was then repeated by controlling the associated personality
variables (i.e., sex, post-depletion and post-TCRT task positive affect; see again Table
5.2). Controlling for the hypothesised covariates yielded a similar pattern of results,
except under high provocation (Table 5.5). Specifically, by controlling positive affect
at both time points, the previously non-significant interaction between mindfulness
induction and depletion on blast intensity became significant, (c3: B = -1.41, SE = .69,
p < .05) when provocation from the opponent was high. The total variance due the
interaction increased from 2.52% to 3.29% (F(1, 113) = 4.24, p < .05), in which
depleted participants with mindfulness induction delivered lower levels of blast
intensity compared to depleted ones without induction. However, there were still no
significant main effects of depletion (c1: B increased from = .52 to .60, p = .08) or
mindfulness induction (c2: B increased from = -.33 to -.45, p = .19) on blast intensity
under high provocations.
161
Taken together, Hypothesis 3 was partly supported such that mindfulness induction
moderated the effect of depletion on blast intensity when participants were being
provoked, but not without provocation, after controlling for covariates. Being male
increased the likelihood of higher levels of blast intensity under no provocation and
low provocation, but the effect of sex diminished on high provocation.
Table 5.5.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, blast intensities as outcome, controlling for covariates
Model testedNo
provocationLow
provocationHigh
provocation
B SE B SE B SE
Depletion (c1) .31 .40 .64* .31 .60 .34Mindfulness induction (c2) -.51 .40 -.36 .31 -.45 .34Depletion x mindfulness (c3) -1.29 .79 -1.29* .62 -1.41* .69R2 on interaction 2.16% 3.31%* 3.29%*
Conditional effect of depletion withoutmindfulness
1.29** .44 1.31** .49
Conditional effect of depletion withmindfulness
-.001 .44 -.10 .48
Covariates:Sex -.85* .39 -.77* .31 - -PANAS PA1 - - - - .55 .41PANAS PA2 - - - - .36 .39Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
PANAS PA1 = Post-depletion positive affect; PANAS PA2 = Post-TCRT task positive affect*p < .05;
**p < .01;
***p < .001
5.4.5.2 Maximum blast latency
Censored survival analyses using Cox regressions were carried out to assess the
extent to which individuals were more likely to administer the maximum blasts as a
function of the experimental conditions. During the TCRT, 17.6% (N = 21)
participants did not deliver any maximum blasts to the opponents. Our proposed
model (mindfulness induction x depletion on maximum blast latency) was not
significant (2 = 5.09, p = .17). Supporting Hypothesis 2b, the likelihood of delivering
the maximum blast to opponent was predicted by the interaction between
mindfulness and depletion (Wald = 4.70, p < .03, CI 95% [.18, .92], exp(B) = .40), but
not by depletion (Hypothesis 1b: Wald = 3.00, p = .08, CI [.94, 2.92], exp(B) = 1.65) or
mindfulness (Wald = 1.40, p = .24, CI [.80, 2.39], exp(B) = 1.39) independently.
162
Depleted participants with who received mindfulness induction were less likely to
deliver maximum blast earlier.
We repeated the censored survival analyses to assess the extent to which depletion
and mindfulness induction influenced maximum blasts latency controlling for the
hypothesised covariates (i.e., sex). The final model was marginally significant (2 =
9.31, p < .054), supporting Hypothesis 3. The point at which participants delivered
the maximum blast was influenced by the interaction of mindfulness and depletion
(Wald = 4.03, p < .05, CI [.19, .98], exp(B) = .43), and sex (Wald = 4.23, p < .05, CI
[1.02 to 2.37], exp(B) = 1.56). Specifically, depleted participants with mindfulness
induction, and females were less likely to deliver maximum blast earlier to the
opponent. There was no main effect of mindfulness (Wald = .40, p = .52, CI [.69,
2.10], exp(B) = 1.20) or depletion (Wald = 2.56, p = .20, CI [.90, 2.82], exp(B) = 1.59).
Given the importance of participant’s sex in predicting the results, we repeated the
censored survival analyses separately by sex. A different effect of depletion and
mindfulness induction on maximum blast latency occurred (Figure 5.4).
Figure 5.4 Sex differences in maximum blast latency. Error bars represent standard error ofmean, asterisk indicates that participants waited significantly longer (p < .01) before
delivering the maximum blast.
For females (n = 45), the final model was significant (2 = 8.88, p < .05). Depleted
females who received mindfulness induction were less likely to deliver the maximum
blast earlier (Wald = 5.77, p < .01, CI [.06, .75], exp(B) = .21). There was also a
significant main effect of mindfulness induction, in which females with mindfulness
163
induction tended to give the maximum blast later (Wald = 7.86, p < .01, CI [1.45,
8.20], exp(B) = 3.45), but no effect of depletion (Wald = 2.62, p = .11, CI [.86, 4.85],
exp(B) = 2.04) on maximum blast latency.
For males (n = 53), the final model was not significant (2 = 6.40, p = .09). Maximum
blasts latency was not predicted by depletion (Wald = .39, p = .53, CI [.60, 2.73],
exp(B) = 1.27), mindfulness (Wald = 2.11, p = .15, CI [.24, 1.24], exp(B) = .55), or
interaction between conditions (Wald = .43, p = .51, CI [.23, 2.08], exp(B) = .69).
5.4.5.3 Correlation between blast intensity and maximum blast latency
For participants who had delivered the maximum blast (N = 98), there were
significant negative correlations between maximum blast latency and blast intensity
across levels of provocation (r = -.57, p <.0001 on no provocation, r = -.52, p < .0001
on low provocation, and r = -.22, p < .05 on high provocation). This finding supported
the conclusion that higher blast intensity was delivered by those who delivered the
maximum blast to the opponent earlier in the TCRT. A similar pattern of results
emerged when we broke down the effect by sex, except for blasts delivered under
high provocation. Specifically, the correlation between maximum blast latency and
blast intensity on high provocation was non-significant for males (r = -19, p = 18),
whereas it was marginally significant for females (r = -.29, p < .052)
5.4.6 Moderation of mindfulness induction on the link between depletion and
indirect aggression
The moderation of mindfulness induction on the link between depletion and each
indirect aggression item was analysed using a bootstrapping method, with 99% bias-
corrected confidence intervals. Hypothesis 2c was not supported in that there was
no significant effect of mindfulness induction on the link between depletion and
indirect aggression in any ratings (Table 5.6). The main impact of depletion on rating
of the opponent’s aggressiveness (Hypothesis 1c) did not survive Bonferroni
corrections (p = .049).
164
Table 5.6.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, rating of opponent as outcome
Model testedAggressive Skilful Competitive Fair Play again
B SE B SE B SE B SE B SE
Depletion (c1) .50* .35 .44 .24 .26 .28 -.30 .29 -.14 .25
Mindfulness induction (c2).13 .25 -
.001.24 -.02 .28 -.27 .29 .31 .25
Depletion x mindfulness(c3)
.07 .50 -.27 .49 -.18 .56 .53 .58 .36 .49
R2 on interaction .02% .26% .09% .71% .45%Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
*p < .05;
**p < .01;
***p < .001
As the opponent ratings did not differ by sex (p = .11 to 61), moderation models
were analysed for males and females together. Sex differences in ratings of indirect
aggression are presented in Table 5.7.
Table 5.7.Sex differences based on rating of opponent
Indirect aggressionMales (n = 52) Females (n = 58)
tM (SD) M (SD)
Aggressive 4.73 (1.46) 4.87 (1.31) t(117) = -.54, p = .59Skilful 4.78 (1.34) 4.92 (1.33) t(117) = -.56, p = .58Competitive 4.14 (1.50) 4.25 (1.56) t(117) = -.41, p = .68Fair 5.05 (1.64) 5.52 (1.49) t(117) = -1.62, p = .11Play again 5.63 (1.43) 5.50 (1.26) t(117) = .52, p = .61*p < .05;
**p < .01;
***p < .001
The moderation of mindfulness on the depletion and ratings of the opponent
aggressiveness, skill, and participants willingness to play against the same opponent
again were repeated by controlling significantly associated personality variables (see
again Table 4.2, no covariates could be proposed for ratings of opponent
competitiveness or fairness). As shown in Table 5.8, no significant effect of
experimental conditions on ratings for these items were found (Hypothesis 3).
Depleted participants continued to rate the opponent as being more “aggressive”
than non-depleted ones, but this effect did not survive Bonferroni corrections. For
covariates that survived Bonferroni corrections, those who were higher in negative
affect following the depletion task rated the opponent as being more “skilful”,
165
whereas those higher in hostility and positive affect after the TCRT task indicated a
greater willingness to take a part in a task with the same opponent again (Table 5.8).
Table 5.8.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, rating of opponent as outcome, controlling for covariates
Model testedAggressive Skilful Play again
B SE B SE B SE
Depletion (c1) .47 .24* .27 .24 -.04 .23Mindfulness induction (c2) .07 .24 .09 .24 .19 .23Depletion x mindfulness (c3) -.18 .49 -.45 .46 .08 .45R2 on interaction .11% .70% .02%Covariates:AQ Physical aggression -.39 .21 -.43 .20* - -AQ Verbal Aggression -.30 .20 - - - -AQ Hostility - - - - .46 .18**
STAR Provocations - - -.32 .16* - -PANAS PA1 .36 .29 - - -.07 .27PANAS NA1 - - .86 .26** - -PANAS PA2 .22 .27 - - .76 .25**
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
5.4.7 Correlation between blast intensity and indirect aggression
No significant correlations were found between blast intensity and ratings of indirect
aggression (p =.17 to .99). An equivalent pattern was also found when we broke
down the correlations by sex.
5.4.8 Moderation of mindfulness induction on the link between depletion and
self-control performance
Table 5.9 shows the changes self-control performance/handgrip stamina prior to the
TCRT task (Change 1) and post-TCRT task (Change 2). Both changes were positively
correlated (r = .62, p < .0001).
166
Table 5.9.Overall change in handgrip performance
Depletion No depletion
Change 1 Change 2 Change 1 Change 2
Mindfulness induction 6.87 -7.30 7.03 -1.07No mindfulness -15.76 -18.65 -3.87 -9.07Note. Higher positive scores indicate better self-control on all measures. Values represent thechanges in mean time that participants squeezed the handgrip pre-TCRT task (Change 1) and post-TCRT task (Change 2).
The hypothesised covariates (i.e., curiosity and decentering at Change 1; sex and SF
at Change 2; see again Table 5.2) were entered into the moderation models (Table
5.10). As expected (Hypothesis 4a), at Change 1, depleted participants who received
mindfulness induction outperformed depleted ones with no induction. The
predicting role of curiosity and decentering was not significant. The moderation of
mindfulness on the depletion and handgrip performance link did not persist after the
TCRT task. Specifically, at Change 2, a better handgrip performance was shown by
non-depleted participants, those who received mindfulness induction, females, and
those higher in SF—thus Hypothesis 4b could not be supported. Sex differences in
handgrip stamina are shown in Figure 5.5.
Table 5.10.Moderation models with bootstrapping method, using depletion as predictor, mindfulness asmoderator, changes in handgrip performance as outcome, controlling for covariates
Model testedChange 1 Change 2
B SE B SE
Depletion (c1) -6.21* 2.86 -7.25** 2.68Mindfulness induction (c2) 15.74*** 2.92 10.35*** 2.68Depletion x mindfulness (c3) 11.90* 5.70 3.93 5.33R2 on interaction 2.73% .36%Conditional effect of depletion without mindfulness -12.21** 4.05Conditional effect of depletion with mindfulness -.31 4.02Covariates:TMS curiosity 2.74 3.31 - -TMS decentering -1.32 3.75 - -STAR Frustrations - - 3.86* 1.69Sex - - 6.38* 2.74Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
Change 1 = mean time differences between handgrip duration at baseline and pre-TCRT task; Change2 = mean time differences between handgrip duration at baseline and post-TCRT task*p < .05;
**p < .01;
***p < .001
167
Figure 5.5. Sex differences in handgrip stamina. Higher positive scores indicate better self-control in Change 1 (pre-TCRT task) and Change 2 (post-TCRT task). Error bars representstandard error of mean, asterisk indicates significant (p < .05) sex differences in handgrip
stamina.
5.5 DISCUSSION
5.5.1 Effect of mindfulness induction on aggressive behaviour following depletion
The current study is a cross-cultural replication of Study 3, using 119 (60 females)
Indonesian university students. Before discussing the effect of experimental
manipulations on aggressive behaviour, we should acknowledge that the absence of
correlation between levels of provocation and blast intensity in the current sample is
quite unusual in the TCRT experiments (e.g., Anderson et al., 2008; Lawrence &
Hutchinson, 2013a, 2013b; also our results from Study 2 and 3). This finding was not
inflated by prior depletion condition (i.e., no differences in overall blast intensities
between depleted and non-depleted participants). Whilst the validity studies of the
Taylor paradigm have been conducted in Western countries (e.g., Giancola & Parrott,
2008), our result indicates the possibility that the TCRT task may be less sensitive to
activate differential levels of provoked aggression in the Indonesian sample.
The lack of sensitivity to the TCRT method was also suggested by the non-significant
correlation between blast intensity and individual differences in sensitivity to
provocations (SP). SP was shown to be related to blast intensity under provoked
aggression in the British sample (also in Study 2 and in Lawrence & Hutchinson,
2013b), but was not related to blast intensity delivered here in the Indonesian
168
sample. Altogether, these notable divergences need to be examined by statistically
comparing the data from British and Indonesian samples directly, which will be
performed in the next chapter.
Crucially, our main analysis showed that mindfulness induction significantly
moderated the link between ego-depletion and blast intensity on low provocation.
This finding reiterates the result from the British sample, suggesting that the benefits
of mindfulness in counteracting the effect of depleted self-control resource on direct
physical aggression may operate similarly across cultures. Controlling for covariates
(i.e., sex and positive affect) provided further support for the moderation model
under high provocation.
As found in the British sample, being Indonesian male also increased the likelihood
of higher levels of blast intensities on no provocation and low provocations, but the
effect of sex diminished on high provocations. This finding is in line with Bettencourt
and Miller’s (1996) meta-analysis in the Western cultures. Bettencourt and Miller
concluded that for females, the differences between aggression under provocation
and neutral conditions are expected to be larger than for males, indicating the
importance of provocation as a moderator of sex differences in aggression. Thus
justification to display aggressive behaviour that is normally held in check by sex role
norms appears to work similarly for females across cultures.
The benefit of mindfulness induction on the maximum blast latency following
depletion was noticeable in particular for females. Specifically, depleted Indonesian
females who received mindfulness induction were less likely to deliver the maximum
blast earlier compared to depleted females with no mindfulness induction. Thus for
females, mindfulness induction reduced physical aggression in both the explicit form
(i.e., blast latency) and the subtle/implicit form (i.e., blast latency); whereas for
males, reductions in aggression due to mindfulness induction occurred mainly in the
explicit form of aggression. Whether this result truly reflects that mindfulness
interventions would be more beneficial for females than for males aggression will
also be elaborated in the next chapter, since in the British female sample,
mindfulness increased the likelihood of delivering the maximum blast earlier.
169
Ratings of indirect aggression (reputation damage of the opponent) did not differ by
sex, similar to the finding from the British participants. Moreover, mindfulness
induction did not moderate the link between depletion and these ratings. Those who
were higher in negative affect following the depletion task rated the opponent as
being more “skilful”, whereas those higher in hostility and positive affect after the
TCRT task indicated a greater willingness to take a part in a task with the same
opponent again. However, these findings were not consistent across all evaluations.
In addition, data from the World Health Organization (WHO, 2008) revealed the
epidemiological profile of suicidal behaviour in Asia typically differs from that of the
Western countries, in which impulsiveness is the major risk factor for suicide in Asia,
whereas suicide in Western countries often involves depression and alcohol abuse.
Therefore, a differential effect of self-harm on aggression is plausible across cultures.
However, as in the British sample, neither direct nor indirect aggression in the
Indonesian sample was related to self-harm, even though self-harm was positively
associated to most of the self-reported measures of aggression. This result implies
that the TCRT task, although performed following depletion condition, may still have
not activated the potential link between self-harm and aggression across cultures.
5.5.2 Effect of mindfulness induction on self-control performance following
depletion
Mirroring the effect of mindfulness induction on direct aggression following
depletion, depleted participants with mindfulness induction also showed a better
self-control performance (assessed by changes in handgrip duration relative to
baseline) than depleted participants with no mindfulness induction when self-control
performance was measured prior to the TCRT task. There was also a positive main
effect of mindfulness induction and a usual main effect of depletion condition on
self-control performance. This strengthens the conclusion of the benefit of
mindfulness induction in the context where individuals lack self-control resource, by
fostering a higher resistance to depletion.
170
While the main effect of mindfulness and depletion persisted when self-control
performance was measured after the experience of provocation during the TCRT, the
moderation of mindfulness did not survive. At this point, the effect of mindfulness
on self-control performance was no longer dependent upon individuals having been
subjected to the ego-depletion condition. A similar result was obtained in the British
sample, whereas in Study 2 we found a crossover interaction between mindfulness
induction and self-control training on self-control performance (i.e., participants
receiving mindfulness induction only or self-control training only showed a better
handgrip power than those who received either both conditions or no condition). It
seems that the short-term effect of brief mindfulness inductions would not hold up
for an extended period of time in particular when people have experienced
provocation. In addition, females and those higher in sensitivity to frustrations
demonstrated increased final self-control performance.
5.5.3 Strengths and limitations of the current study
The current study provides exploration of Western’s operationalisation of
mindfulness, self-control, aggression, and self-harm in the context of an Asian
culture, particularly Indonesian. As found in the British sample, the 15-min
mindfulness breathing exercise increased mindfulness state of decentering, but not
curiosity. The attention control task was also effective to induce depletion (higher
ratings in attention control and difficulty of the task). Further, no changes were
found in postdepletion mood, dovetail with past research demonstrating that
depletion manipulations have no impact on mood (e.g., Baumeister et al., 1998;
Muraven et al., 1998; Stucke & Baumeister, 2006). Thus in general, the conditions for
the mindfulness induction and depletion task were comparable across cultures.
Crucially, replicating the outcomes from British sample, the impact of mindfulness
induction was established in the Indonesian sample in terms of reduced in blast
intensity following depletion.
171
Due to time and resources constraints, we did not precisely follow the
recommended guidelines for cross-cultural adaptation of the translated instruments
(e.g., Beaton, Bombardier, Guillemin, & Ferraz, 2000; Gjersing, Caplehorn, & Clausen,
2010). After the initial translation and synthesis of the measures by two
postgraduate Indonesian students, the back translation of the questionnaires was
provided through a commercial translation service instead of using translators with
English mother tongue. Although the adapted measures were then pre-tested (with
2 pilot participants), no additional testing on the psychometric properties of the
questionnaires was conducted. While the adapted measures might have reflected
the content validity of the original measures, we could not provide any information
with regards to the construct validity, reliability, or item response patterns.
Moreover, it would be difficult for this study to raise definitive conclusions about the
role of provocations on aggression across cultures, or the efficacy of the Taylor
paradigm outside of US and UK/European samples, due to the absence of association
participants’ responses to aggression and provocation level from the opponent. This
issue will be detailed in the next chapter.
5.5.4 Conclusions and next steps
Consistent with the results from the British sample, the current study showed that
the benefits of mindfulness on direct aggression may be more salient under depleted
condition than under full self-control resources. Relatedly, the maximum blast
latency seems to capture females’ aggression, and females outperformed males in
self-control performance. No influence of experimental conditions was found on
indirect aggression (reputation damage to the opponent). The following chapter will
focus on the strength of the effect of mindfulness induction across cultures.
172
CHAPTER SIX
Cross-cultural comparison of the effect of mindfulness induction on aggressive
behaviour following depletion (British vs. Indonesian sample)
6.1 INTRODUCTION
The current chapter examines cross-cultural comparison between the native British
(Study 3) and Indonesian participants (Study 4). The comparison is made based on
cultural differences in (i) the strength of the effect of mindfulness induction on
experimental outcomes (i.e., blast intensity, maximum blast latency, indirect
aggression, and self-control performance) under depleted self-control resources, and
(ii) personality variables.
6.1.1 Cultural differences in the effect of mindfulness induction on aggressive
behaviour following depletion
Past research employing the Taylor Competitive Reaction Time (TCRT: Taylor, 1967)
task has typically demonstrated significant positive effects of levels of provocation
from the opponent on participants’ levels of blast intensity (e.g., Anderson et al.,
2008; DeWall et al., 2010; Konrath et al., 2006; Lawrence & Hutchinson, 2013a,
2013b). This standard provocation effect was shown in the British sample, but not in
the Indonesian sample. As this is of the first study we know to employ the TCRT task
in Indonesia, it is difficult to conclude whether the Indonesian subjects were simply
less responsive to physical provocation (i.e., loud blasts), or whether they were more
behaviourally aggressive per se—and so operating at ceiling type levels for them. The
current analysis directly examines the differences between Indonesian and British
participants’ responses to aggression under provocation and at baseline level.
173
The key analysis in this chapter concerns whether mindfulness as a moderator of the
depletion and aggression link varies as a function of culture. In other words, we are
interested in testing whether culture moderates the moderation of mindfulness on
the depletion and aggression relationship. As presented in the two previous
chapters, under low provocation trials, both depleted Indonesian and British
participants who received mindfulness induction displayed lower levels of blast
intensity as opposed to their depleted counterparts with no mindfulness induction.
This effect was also found under no provocation for the British, but not for the
Indonesians. The effect for the Indonesians occurred on high provocation after
controlling for positive affect (although positive affect did not predict blast
intensity). While we expect the benefit of mindfulness to hold regardless of the
cultural context, the effect might be stronger in the Indonesians as they are more
familiar to the concepts of mindfulness, or in the British as the novelty factor of this
technique may produce an increasing effect. On the other hand, the main effect of
cultures may occur in direct aggression. As the general collectivism culture in
Indonesia strongly prohibits display of aggression (Koentjaraningrat, 1985; Magnis-
Suseno, 1997), the Indonesian sample may display less blast intensities in general.
Previous analyses showed that for both cultures, no support was found for the
proposed depletion x mindfulness model on maximum blast latency (the delay
duration before the maximum aggressive response was delivered). However, after
the inclusion of participants’ sex, depleted Indonesian participants with mindfulness
induction, and Indonesian females were less likely to deliver maximum blast to the
opponent. The opposing effect was evident for British females. These findings
suggest the necessity to include participants’ sex, and that mindfulness intervention
may be more effective to undermine the Indonesian females’ aggression in terms of
maximum blast latency than British females. So, like blast duration (Giancola &
Parrott, 2008), blast latency may be operating as an indirect measure of aggression
for British females, but more akin to direct aggression for Indonesian females.
174
Previous analyses also showed the lack of efficacy of mindfulness induction in
counteracting the effect of depletion on indirect aggression (anonymous reputation
damage to the opponent) for both the British and Indonesian samples. Relatedly, it
has been shown elsewhere that the presence of collectivism value could limit the
exhibition of indirect aggression (Forbes et al., 2009), but may also increase the
tendency to adopt this method (French et al., 2002). Therefore, cultural differences
in the effectiveness of mindfulness induction as well as the main effect of culture on
indirect aggression may be less observable in the current analysis.
It should be noted, however, that sex differences occurred in French et al.’s (2002)
study comparing descriptions of disliked peers by the Indonesian and US youth (in
Hofstede’s, 2010 map of individualism, the US receives a score 91, almost equal to
the UK). In both countries, boys associated disliked peers with their exhibition of
physical aggression by these peers, whereas girls associated these peers with
relational aggression more often than boys. According to French et al., a disliked
peer could be described as engaging in certain behaviour because the behaviour
occurs frequently—thus confirming a typical sex differences in direct and indirect
aggression across cultures; or conversely, because the behaviour is so infrequent
that it is particularly salient or socially unacceptable. In other studies, however, sex
was shown as a weak contributor to indirect aggression amongst college students in
the US (individualist) and China (collectivist: Forbes et al., 2009), suggesting that
cross-culturally sex differences in indirect aggression may be less observable in
adults than in younger population. Thus in the current analysis, no strong prediction
is made with regards to sex differences in indirect aggression a function of culture.
6.1.2 Cultural differences in the effect of mindfulness induction on self-control
performance following depletion
A noticeable cultural difference may occur on the link between depletion and self-
control performance (handgrip stamina). Apart from past evidence that collectivism
175
cultures promote a better resistance to ego-depletion (Balliet & Joreiman, 2010;
Seeley & Gardner, 2003; see also Baumeister, Vohs, et al., 2007), our previous
analyses showed that mindfulness induction did increase self-control performance
following depletion for the Indonesian sample, but not for the British. Therefore, a
stronger moderation effect of mindfulness is expected to favour the Indonesians.
Consistent with the above prediction, the main effect of culture on the link between
depletion and self-control performance is also expected to be stronger in the
Indonesian than in the British sample. In addition, the previous chapters consistently
revealed that females in both cultures were more able to maintain their baseline
handgrip performance than counterparts male. Thus it may be that cultural
differences in self-control performance would be less pronounced for females.
6.1.3 Cultural differences in personality variables
Individual differences in sensitivity to provocations (SP) was positively associated to
blast intensity under provoked aggression in the British sample (also in Study 2 and
in Lawrence & Hutchinson, 2013b), but not in the Indonesian sample. The current
analysis allows us to examine the extent to which provocations and frustrations
sensitivity (SF) is similar in both cultures, since no studies examining the reliability
and validity of the SP or SF in a non-Western context have been conducted to date.
Cultural differences in the overt or behavioural component of aggression have been
reported elsewhere (e.g., Forbes et al., 2009; Ramirez et al., 2001). This implies that
the Indonesian people may possess fewer propensities to behave aggressively,
particularly in terms of physical and verbal aggression. This could also mean that the
Indonesians may be more sensitive to signals of aggression than the British people.
Alternatively, even though females are commonly more socialised to inhibit their
exhibition of aggression, they generally score the same as their male counterparts on
measures of both SP and SF (Lawrence, 2006), as well as on anger and hostility
(Archer, 2004). Therefore, in the same way, the Indonesian sample, while having
176
strong social sanctions to limit the exhibition of aggression behaviour, may score
similarly in terms of trait aggression, anger, hostility, SP and SF, but may score lower
in trait physical and verbal aggression.
The current analysis also compares cultural differences in the individual’s propensity
to be mindful and self-controlled. Compared to the British, the Indonesian people,
particularly Javanese Indonesians are already accustomed to mindful ways of
thinking in daily life (Koentjaraningrat, 1989). Thus a higher level of trait mindfulness
is expected for the Indonesians. However, cultural differences in trait self-control
would be less expected, because self-controlled regulation of personal goals is
prevalent in individualistic societies (Brown et al., 2007a), but compliance with
societal norm is promoted in the collectivists (Hofstede, 1980, 2010; Triandis, 1995).
In addition, we also analyse cultural differences in the prevalence of self-harm.
Suicidal behaviour in developing countries in Asia has been typically underreported,
due to various religious, legal and cultural constraints (Kahn, 2005; Vijayakumar,
2004). However, a recent study with university students in Indonesia reported a 38%
lifetime prevalence of self-harm, comparable with Gratz’s (2001) study using
students in the US (Tresno, Ito, & Mearns, 2012). Thus we proposed no specific
prediction with regards to the prevalence of self-harm across the current samples.
It should be underlined that the assessment of the role of individual differences in
mindfulness, self-control, aggression, self-harm, SP, and SF as covariate has been
conducted separately in Chapter 4 and 5, so the current analysis focuses only on
cultural differences in the aforementioned traits.
6.2 HYPOTHESES
The cross-cultural analyses examine the following hypothesis:
1. For personality traits across cultures:
177
a. Indonesians will show higher levels of trait mindfulness than the British.
No strong prediction will be made for cultural differences in trait self-
control.
b. Indonesians will show lower levels of trait physical and verbal aggression
than the British. No strong prediction will be made for cultural differences
in general trait aggression, anger, and hostility.
c. No strong prediction will be made for cultural differences in sensitivity to
provocations (SP) and frustrations (SF).
d. No strong prediction will be made for cultural differences in self-harm.
2. For direct aggression in terms of blast intensity delivered to the opponent:
a. Compared to British participants, Indonesians will deliver lower levels of
blast intensity. Blast intensity is measured across three provocation levels
(i.e., no provocation, low provocation, high provocation).
b. The moderation of mindfulness on the depletion and blast intensity link
will be expected to hold regardless of culture, but could either be
stronger for Indonesians participants or for the British.
3. For maximum blast latency:
a. No cultural differences will be expected in the maximum blast latency.
Cultural differences may occur with the inclusion of participants’ sex, in
which Indonesian females will wait longer before delivering the maximum
blasts to the opponent than the British counterparts.
b. No specific prediction will be made for cultural differences in the
moderation of mindfulness on the depletion and maximum blast latency
link. After the inclusion of participants’ sex, cultural differences are
expected to favour depleted Indonesians than the British participants.
4. For indirect aggression (anonymous reputation damage to the opponent):
a. No cultural differences will be expected in indirect aggression.
Additionally, no specific prediction is made for cultural differences in
indirect aggression across sexes.
178
b. No specific prediction will be made for cultural differences in the
moderation of mindfulness induction on the depletion and indirect
aggression link.
5. For self-control performance (changes in handgrip stamina):
a. Compared to British participants, Indonesians will be more able to
maintain their self-control performance. Cultural differences in self-
control performance may be more observable in males than in females.
b. Cultural differences in the moderation of mindfulness induction on the
depletion and self-control performance link will be expected to favour
Indonesians than British participants.
6.3 METHODS
6.3.1 Participants
The final cross-cultural sample consisted of 119 Indonesian and 110 native British
university students. The numbers of males and females in each experimental
condition (2 [ego-depletion vs. no depletion] x 2 [mindfulness induction vs. no
induction] are shown in Table 6.1 (see section 4.3.4, Chapter 4 for details of the
experimental manipulations).
Table 6.1.Frequency distributions for cross-cultural analysis
DepletionMindfulness induction
Mindfulness No mindfulness
Depletion Indonesian 30 (15 M, 15 F) 29 (14 M, 15 F)
British 28 (13 M, 15 F) 27 (13 M, 14 F)
No depletion Indonesian 30 (15 M, 15 F) 30 (15 M, 15 F)
British 27 (14 M, 13 F) 28 (12 M, 16 F)
Note. M = Males; F = Females
179
6.3.2 Data analysis
A series of independent t-tests, one-way ANOVAs, and Chi-square tests was
performed to test cross-cultural differences in personality variables (Hypothesis 1a,
1b, 1c, and 1d). The moderation of mindfulness and culture on the link between
depletion and aggressive behaviour and self-control performance was tested with
bootstrapping method, using model “number 3” in A. F. Hayes’s (2012a) PROCESS
macro for three-way interaction (moderated moderation).
Figure 6.1. Conceptual and statistical models for moderated moderation. X = predictor(depletion), Y = outcome (aggression or self-control performance), M = first moderator(mindfulness), W = second moderator (culture). c1, c2, c3, c4, c5, c6, c7 = unstandardisedregression coefficient. Reprinted from A. F. Hayes (2012b, p.33).
In a moderated moderation model, the effect of X on Y depends multiplicatively on
M and W (Figure 6.1). The model is proposed mathematically as: Y = i + c1X + c2M +
c3W + c4XM + c5XW + c6MW + c7XMW + ey. Three-way interaction is present if c7 is
statistically different from zero. Alternatively, the analysis can focus on the
conditional two-way interaction between X and M as moderated by W = c4 + c7W.
This can be understood by selecting values of W and testing whether the conditional
interaction between X and M is statistically different from zero at those values.
180
In the current analysis, the moderated moderation model allows us to test whether
the conditional two-way interaction between mindfulness and depletion would be
statistically different from zero at the value of culture, on (i) blast intensity on each
provocation levels (Hypothesis 2b), (ii) maximum blast latency (Hypothesis 3b), (iii)
anonymous reputation damage of the opponent (Hypothesis 4b), and (iv) change in
handgrip stamina pre- and post-TCRT task (Hypothesis 5b) as the outcomes.
Additionally, this model also predicts the main effect of culture on the
aforementioned outcomes (Hypothesis 2a, 3a, 4a, and 5a).
6.4 RESULTS
6.4.1 Cultural differences in personality variables
Table 6.2.Cross-cultural differences in personality measures
Personality measuresIndonesian British
tM (SD) M (SD)
MAAS 3.93 (.57) 3.72 (.70) t(227) = 2.53**
SCS 3.14 (.46) 3.05 (.56) t(212.28) = 1.25, p = .21Total AQ 2.76 (.45) 2.43 (.57) t(207.70) = 4.75***
AQ Physical 2.14 (.61) 2.05 (.83) t(199.58) = .95, p = .34AQ Verbal 3.14 (.63) 2.88 (84) t(201.87) = 2.67**
AQ Anger 2.67 (.76) 2.22 (.72) t(227) = 4.55***
AQ Hostility 3.09 (.64) 2.56 (.79) t(210.90) = 5.28***
STAR Provocations 2.92 (.78) 2.88 (.78) t(227) = .30, p = .77STAR Frustrations 2.76 (.82). 2.54 (.76) t(227) = 2.15*
Note. MAAS = Mindful Attention Awareness Scale; SCS = Self-Control Scale; AQ = Aggression
Questionnaire; STAR = Situational Triggers of Aggressive Responses*p < .05;
**p < .01;
***p < .001
Only one British participant reported practicing mindfulness (on a monthly basis),
whereas none of the Indonesians reported having encounters with these
experiences. Table 6.2 depicts the cross-cultural comparison in personality
measures. As expected, compared to British participants (N = 110), the Indonesians
(N = 119) scored significantly higher on trait mindfulness, but not on trait self-control
181
(Hypothesis 1a). No support was found for Hypothesis 1b, as the Indonesians scored
higher in verbal aggression, but similar to the British in physical aggression. In fact,
the Indonesians also scored higher in general trait aggression, anger, and hostility.
Hypothesis 1c was partially supported in that equal level of sensitivity to provocation
(SP) was reported between cultures, although the Indonesian sample was higher in
sensitivity to frustrations (SF). Supporting Hypothesis 1d, self-harm did not vary as a
function of culture (2(1) = 3.34, p = .07), although the rate of self-harm in the
Indonesians tend to be higher (n Indonesian = 22 [66.7% of total]; n British = 11
[33.3% of the total]). In short, compared to the British, the Indonesians were more
mindful, equally self-controlled, and more aggressive in feelings.
Additionally, sex differences occurred for physical aggression, (F(1,225) = 21.29, p <
.0001; M males = 2.31, SD = .80 vs. M females = 1.89, SD =.57), anger (F(1,225) =
4.97, p < .05; M males = 2.35, SD = .76 vs. M females = 2.56, SD =.77), and SF
(F(1,225) = 12.37, p < .001; M males = 2.47, SD = .78 vs. M females = 2.83, SD =.78).
Comparable rates of self-harm were shown between males and females (2(1) = .57,
p = .45; n males = 14 [42.4% of the total]; n females = 19 [57.6% of the total]). No
significant interactions were found between sex and culture on the trait measures (p
= .09 to .99). Thus males and females across cultures reported a comparable pattern
of results with regards to the tendency to harm themselves and others.
6.4.2 Cultural differences in the moderation of mindfulness induction on the link
between depletion and aggressive behaviour
6.4.2.1 Levels of blast intensity
A bootstrapping method with 95% bias-corrected confidence intervals (based on
5,000 bootstrap resamples, N = 229) was employed to test the moderated
moderation model. All predictors were mean-centered. Results in Table 6.3 show
comparable effectiveness of the mindfulness intervention for the depleted group
across both samples (Hypothesis 2b: c7, p = .61 to .98). Regardless of culture,
182
mindfulness moderated the link between depletion and blast intensities in all
provocation levels (c4), in which both depleted Indonesian and British participants
with mindfulness induction delivered lower levels of blast intensity than depleted
ones with no mindfulness induction. Under no provocation and low provocation,
lower blast intensity was also delivered by mindfulness induction (c2) and non-
depleted (c1) participants. Contrary to Hypothesis 2a (c3), Indonesians delivered
higher blast intensities than British across all provocation levels (see Figure 6.2).
Table 6.3.Moderation models with bootstrapping method, using depletion as predictor, mindfulnessand culture as moderator, blast intensity as outcome
Model testedNo
provocationLow
provocationHigh
provocation
B SE B SE B SE
Mindfulness induction (c2) -.72** .30 -.48* .24 -.47 .41Depletion (c1) .75** .30 .63** .24 .47 .41Depletion x mindfulness (c4) -1.58** .60 -1.41** .49 -1.24* .54Culture (c3) -3.00*** .30 -1.93*** .24 -1.24*** .27Depletion x culture (c5) .93 .60 -.02 .49 -.10 .54Mindfulness x culture (c6) -.44 .60 -.27 .49 -.28 .54Depletion x mindfulness x culture (c7) -.62 1.20 -.27 .97 -.02 1.07R2 on three-way interaction .08% .03% 0%Conditional effect of mindfulness xdepletion for Indonesians
-1.28 .84 -1.28 .67 -1.23 .74
Conditional effect of mindfulnessx depletion for British
-1.90* .87 -1.55* .70 -1.25 .77
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
In addition, both Indonesian males and females delivered higher blast intensities
across all provocation levels relative to their British counterparts (Table 6.4). Thus
cultural differences in direct aggression were consistently shown across sexes.
Table 6.4.Means for blast intensity across culture and sex
Blast intensity SexIndonesian British
t
M SD M SD
No provocation blast Male 6.03 2.05 3.06 2.89 t(90.56) = 6.18***
Female 5.20 2.30 2.21 2.03 t(116) = 7.74***
183
Blast intensity SexIndonesian British
t
M SD M SD
Total 5.61 2.21 2.61 2.50 t(227) = 9.65***
Low provocation blast Male 5.70 1.83 3.81 2.24 t(109) = 4.87***
Female 4.94 1.67 3.00 1.69 t(116) = 6.29***
Total 5.31 1.78 3.38 2.00 t(227) = 7.73***
High provocation blast Male 5.49 2.17 4.35 2.29 t(109) = 2.68**
Female 5.26 1.68 3.94 2.06 t(116) = 3.80***
Total 5.37 1.94 4.14 2.17 t(227) = 4.55***
*p < .05;
**p < .01;
***p < .001
Figure 6.2. Cross-cultural comparisons of blast intensity based on experimental conditions(mindfulness and depletion). Error bars indicate standard error of mean. Scale goes up to 8
as a maximum.
184
6.4.2.2 Maximum blasts latency
During the TCRT, the maximum blasts were delivered by 175 participants (98
Indonesian, 77 British). To support our prediction that no cultural differences would
be shown in the moderation of mindfulness on the depletion and maximum blast
latency link, we firstly carried out a censored survival analyses using Cox regressions.
As expected, this proposed model was not significant (2 = 4.80, p = .44). The
likelihood of delivering maximum blast earlier was not predicted by interaction
between mindfulness x depletion x culture (Hypothesis 3b: Wald = 2.81, p = .09, CI
[.25, 1.12], exp(B) = .53). The main effect of culture (Hypothesis 3a: Wald = 2.48, p =
.12, CI = [.93, 1.87], exp(B) = 1.32), mindfulness (Wald = .43, p = .52, CI [.76, 1.73],
exp(B) = 1.15), and depletion (Wald = 1.28, p = .26, CI [.83, 1.96]), exp(B) = 1.28) as
well as the interaction of mindfulness x depletion (Wald = .002, p = .96, CI [.48, 2.01],
exp(B) = .98) on maximum blast latency were all non-significant.
To support the importance of sex differences on maximum blast latency, we
repeated the censored survival analysis by including participants’ sex into the model.
The final model for depletion, mindfulness, culture, and sex on maximum blast
latency was significant (2 = 17.40, p < .01), with a significant three-way interaction
between mindfulness x depletion x culture (Wald = 4.50, p < .05, CI [.20, .94], exp(B)
= .44). Contrary to Hypothesis 3b, however, depleted British participants with
mindfulness induction were less likely to deliver the maximum blast earlier than
their Indonesian counterparts. Relatedly, the main effect of sex was also significant
(Wald = 12.44, p < .0001, CI [1.29, 2.41], exp(B) = 1.15), but in the reverse direction
to our prediction (Hypothesis 3a). Specifically, compared to females, males in both
cultures waited longer before delivering the maximum blast. In addition, maximum
blast latency was not predicted by the main effect of culture (Wald = 2.76, p = .10, CI
[.95, 1.90], exp(B) = 1.34), mindfulness (Wald = .00, p = .99, CI [.67, 1.51], exp(B) =
1.00), and depletion (Wald = 1.20, p = .26, CI [.78, 1.84], exp(B) = 1.20), or by the
interaction between mindfulness and depletion (Wald = .33, p = .57, CI [.60, 2.57],
185
exp(B) = 1.24). The proposed model was not significant when males (2 = 5.73, p =
.33) and females (2 = 4.92, p = .43) were analysed separately.
6.4.2.3 Indirect aggression
Table 6.5.Moderation models with bootstrapping method, using depletion as predictor, mindfulnessand culture as moderator, rating of opponent as outcome
Model testedAggressive Skilful Competitive Fair Play again
B SE B SE B SE B SE B SE
Mindfulness induction(c2)
.04 .17 .03 .17 .07 .18 -.16 .21 .28 .19
Depletion (c1) .36 .17* .18 .17 .18 .18 .04 .21 -.09 .19
Depletion x mindfulness(c4)
.05 .35 -.26 .34 -.11 .37 .22 .43 -.16 .39
Culture (c3) -.70*** .17 .05 .17 .91*** .18 -.72*** .21 -.20 .19
Depletion x culture (c5) -.30 .35 -.53 .34 -.17 .37 .70 .43 .12 .39
Mindfulness x culture(c6)
-.19 .35 .06 .34 .18 .37 .23 .43 -.07 .39
Depletion x mindfulnessx culture (c7)
-.04 .70 .02 .68 .14 .74 -.64 .86 -1.08 .77
R2 on three-wayinteraction
0% 0% .01% .24% .86%
Conditional effect ofmindfulness x depletionfor Indonesians
.07 .48 -.27 .47 -.18 .51 .53 .59 .36 .54
Conditional effect ofmindfulnessx depletion for British
.03 .50 -.25 .49 -.04 .53 -.11 .62 -.72 .56
Note. B = unstandardised regression coefficient; SE = standard error; R2
= variance increase*p < .05;
**p < .01;
***p < .001
We used bootstrapping analysis at 99% bias-corrected confidence intervals to
survive Bonferroni corrections (p < .01). As expected, results in Table 6.5 shows no
significant mindfulness x depletion x culture interaction on any of indirect aggression
ratings (Hypothesis 4b: c7, p = .17 to .97). Corroborating the results from the two
previous chapters, conditional two-way effect of mindfulness x depletion (c7) was
not significant for Indonesians (p = .37 to .89), or for British participants (p = .20 to
186
.95). The main positive effect of depletion occurred on rating of the opponent’s
aggressiveness (c1), however, this effect was not strong enough to survive Bonferroni
corrections (p < .041).
We did not propose specific prediction for the main effect of culture on indirect
aggression (Hypothesis 4a). However, culture differences occurred in ratings of the
opponent’s aggressiveness, competitiveness, and fairness (c3). As shown in Table 6.6,
the Indonesians rated the opponent as being more aggressive, but also more fair and
less competitive. No significant cultural differences were found in rating of the
opponent’s skill or in the willingness to play with the same opponent again. As an
additional testing for Hypothesis 4a, a comparable pattern of results in rating of
indirect aggression was obtained across sexes. Specifically, both Indonesian males
and females rated the opponent as being more aggressive but less competitive than
their British counterparts, but Indonesian females also rated the opponent as being
more fair. Taken together, it can be concluded that Indonesian males and females
showed comparable indirect aggression ratings with their British counterparts.
Table 6.6.Means for rating of opponent across culture and sex
Indirect aggression SexIndonesian British
tM SD M SD
Aggressive Male 4.73 1.46 3.94 1.31 t(109) = 2.98**
Female 4.87 1.31 4.24 1.16 t(116) = 2.75**
Total 4.80 1.38 4.10 1.23 t(227) = 4.02***
Skilful Male 4.78 1.34 4.75 1.37 t(109) = .12, p = .91
Female 4.92 1.33 5.03 1.08 t(116) = -.53, p = .60
Total 4.85 1.33 4.90 1.23 t(227) = -.30, p = .76
Competitive Male 4.14 1.49 4.87 1.31 t(109) = -2.72**
Female 4.25 1.56 5.31 1.10 t(106.06) = -4.29***
Total 4.19 1.52 5.10 1.22 t(222.64) = -5.00***
Fair Male 5.05 1.64 4.40 1.87 t(109) = 1.94, p = .06
Female 5.52 1.49 4.71 1.41 t(116) = 3.03**
Total 5.29 1.58 4.56 1.65 t(227) = 3.39***
Play again Male 5.63 1.43 5.46 1.70 t(109) = .56, p = .58
Female 5.50 1.26 5.26 1.46 t(116) = .97, p = .34***
187
Indirect aggression SexIndonesian British
tM SD M SD
Total 5.56 1.34 5.35 1.57 t(227) = 1.08, p = .28*p < .05;
**p < .01;
***p < .001
6.4.3 Cultural differences in the moderation of mindfulness induction on the link
between depletion and self-control performance
Changes in self-control performance were calculated by the mean differences
between handgrip stamina at baseline and prior to the TCRT task (Change 1) and
post-TCRT task (Change 2). The result of the moderated moderation analysis is
presented in Table 6.7. Contrary to Hypothesis 5b, there was no significant three-
way interaction between mindfulness x depletion x culture at Change 1, (c7: B = -
8.36, SE = 9.05, p = .09) or Change 2 (c7: B = -7.25, SE = 8.67, p = .40). A better
handgrip performance at both time points was shown by those who received
mindfulness induction or by non-depleted participants.
Table 6.7.Moderation models with bootstrapping method, using depletion as predictor, mindfulnessand culture as moderator, changes in handgrip performance as outcome
Model testedChange 1 Change 2
B SE B SE
Mindfulness induction (c2) 13.12*** 2.26 8.92*** 2.17Depletion (c1) -6.18** 2.26 -4.95* 2.17Depletion x mindfulness (c4) 7.71 4.52 -.13 .98Culture (c3) -7.87*** 2.26 .55 2.17Depletion x culture (c5) -.37 4.53 6.14 4.34Mindfulness x culture (c6) -7.54 4.53 -1.56 4.43Depletion x mindfulness x culture (c7) -8.36 9.05 -7.25 8.67R2 on three-way interaction .30% .28%Conditional effect of mindfulness x depletion forIndonesians
11.73 6.27 3.36 6.01
Conditional effect of mindfulness x depletion for British -3.37 6.53 -3.89 6.25Note. B = unstandardised regression coefficient; SE = standard error; R
2= variance increase
Change 1 = mean time differences between handgrip duration at baseline and pre-TCRT task; Change2 = mean time differences between handgrip duration at baseline and post-TCRT task*p < .05;
**p < .01;
***p < .001
188
As expected (Hypothesis 5a), higher self-control performance at Change 1 was also
influence by culture (c3). Specifically, as can be seen in Table 6.8, regardless of
experimental conditions, Indonesian participants were more able to maintain their
handgrip performance than British participants at Change 1, although comparable
performance was shown at Change 2. The main effect of culture at Change 1
(Indonesians outperformed British) was particularly pronounced for males, but not
for females, thus lending an additional support for Hypothesis 5a.
Table 6.8.Overall change in handgrip performance across culture and sex
Blast intensity SexIndonesian British
t
M SE M SE
Change 1 Male -3.68 2.50 -16.25 3.12 t(109) = 3.17**
Female 1.02 2.15 -3.02 1.68 t(110.48) = 1.48, p = .14
Total -1.31 1.66 -9.27 1.83 t(227) = 3.24***
Change 2 Male -12.80 2.24 -12.87 3.10 t(109) = .02, p = .98
Female -5.15 1.89 -4.53 1.47 t(110.41) = -.26, p = .80
Total -8.94 1.48 -8.47 1.70 t(227) = .84, p = .84
Note. Higher positive scores indicate better self-control on all measures. Values represent thechanges in mean time that participants squeezed the handgrip pre-TCRT task (Change 1) and post-TCRT task (Change 2).*p < .05;
**p < .01;
***p < .001
As a whole, the summary of findings from the current cross-cultural analysis, as well
as from each sample is presented in Appendix 6.1.
6.5 DISCUSSION
6.5.1 Cultural differences in the effect of mindfulness induction on aggressive
behaviour following depletion
The current analysis sought to compare the possible cultural differences in the
effectiveness of mindfulness induction to counteract the negative impact of
depletion on aggressive behaviour, using the results from Chapter 4 and 5.
189
Supporting our prediction, mindfulness induction reduced blast intensities across all
provocation levels for depleted Indonesians to the same extent it affected their
British counterparts. Thus regardless of the influences of culture in shaping
individual’s personality (discussed in the latter section), the current analysis revealed
that the benefit of mindfulness induction on the reductions of aggressive behaviour
under depleted self-control resources is likely to be similar across cultures.
Additionally, the main effect of mindfulness induction and depletion condition on
direct aggression was also similar across cultures. Specifically, under no provocation
and low provocation, depleted Indonesian and British participants delivered higher
blast intensity in the adapted TCRT task than non-depleted ones, whereas those who
received mindfulness induction delivered lower blast intensity than those who did
not receive mindfulness induction. In Lawrence and Hutchinson’s study (2013a; see
also Anderson et al., 2008), participants who were asked to deal with an opponent
that initially refrained from behaving aggressively showed less aggressive responses
in subsequent provocation procedure, compared to participants with an initially
aggressive opponent. As the levels of provocation increased, the impact of the
opponent’s initial behaviours became less significant.
In the same way, the independent role of mindfulness and depletion conditions
across cultures may be less observable as people encounter more severe offences
(i.e., no significant effect of each condition on participant’s blast intensity under high
provocation). Crucially, however, we found that the joint effect of mindfulness and
depletion on participant’s blast intensity persisted even following high provocation
from the opponent. This highlights our conclusion that the cross-cultural benefit of
mindfulness on aggressive behaviour would be salient when individual’s self-control
resource has been depleted, even when the experience of provocation is severe.
The moderation of mindfulness induction on the link between depletion and
maximum blast latency was also equivalent for Indonesian and British participants.
When participants’ sex was accounted in the model, the impact of culture became
significant. Even so, contrary to our prediction, the likelihood of delivering the
190
maximum aggressive responses decreased for British participants with mindfulness
induction compared to their Indonesian counterparts. Although speculative, it is
plausible that the novelty factor of mindfulness interventions may have produce an
additional effect for the British, by delaying their maximum aggressive response
while at the same time reducing their intensity of physical aggression (i.e., blast
intensity levels). The UK has been described as a curious nation, with a low score on
Hofstede et al.’s (2010) uncertainty avoidance dimension (i.e., the extent to which
cultures seek to minimise feeling threatened by ambiguous or unknown situations).
For indirect aggression, the three-way interaction between mindfulness x depletion x
culture was not significant. Given that similar results were obtained from the
separate analyses of the British and Indonesian participants, it is likely that any
potential effect that mindfulness may have on attenuating the impact of depletion is
not observable on indirect aggression.
6.5.2 Cultural differences in the effect of mindfulness induction on self-control
performance following depletion
In the two previous chapters, mindfulness induction moderated the link between
depletion and self-control performance (handgrip stamina) prior to the TCRT-task for
the Indonesians, but not for British participants. Collectivism cultural background has
also been suggested as a possible moderator of ego-depletion (Baumeister, Vohs, et
al., 2007; Balliet & Joreiman, 2010; Seeley & Gardner, 2003). However, the current
analysis revealed that moderating effect of mindfulness on self-control performance
following depletion was similar across cultures. Specifically, only those who received
mindfulness induction and non-depleted participants showed a better self-control
performance than their counterparts. This indicates that despite cultural differences,
the positive independent effect of mindfulness induction and the negative effect of
depletion condition affected self-control performance to the same extent.
191
6.5.3 Cultural differences in personality variables
In the current samples, none of the Indonesian participants and only one British
participant reported having formal experience with mindfulness practices (defined as
“a specific meditational practice in directing attention without making any
judgment”). The Indonesian participants, however, scored higher on trait
mindfulness compared to the British, suggesting that they might have been more
engaged more with the application of mindfulness in daily life (Koentjaraningrat,
1985), without necessarily having experience of formal mindfulness interventions.
This provided an empirical basis for our assumption that those in Indonesia may be
more mindful.
As expected, both samples showed equal levels of trait self-control. Even so, the
underlying reasons for this cross-cultural similarity would probably differ. The British
sample may see self-controlled regulation as means to facilitate the achievement of
personal goals (Brown et al., 2007a), whereas the Indonesian sample may be
motivated to practice self-control because they want to comply with societal norm.
Both reference values could enact as a standard for engaging in self-controlled
regulation (Carver & Scheier, 1982, 2011; Baumeister, Schmeichel, et al., 2007).
While individuals from collectivistic cultures typically reported lower levels of
behavioural component of aggression (e.g., Bergeron & Schneider, 2005; Forbes et
al., 2009; Ramirez et al., 2001), a comparable level of self-reported physical
aggression was found in our sample. The Indonesian participants scored significantly
higher in other subscales of aggression (i.e., verbal aggression, anger, hostility). Their
levels of sensitivity to frustrations (SF) were also higher, but they were equally
sensitive to provocations (SP) with the British participants. Given that trait physical
aggression has been demonstrated as the strongest predictor for direct aggression in
the TCRT task (Giancola & Parrott, 2008), and that SP typically predicted provoked
aggression (Lawrence, 2013b; also our results from Study 2), our result implies that
the Indonesian sample may be equally aggressive with the British in the TCRT task.
As discussed in the next section, this was not the case in the current analysis, as the
192
Indonesian delivered significantly higher blast intensities in all provocation levels
compared to the British sample.
Additionally, with regards to the potential link between aggression and self-harm, no
significant culture or sex differences between self-harmers and non self-harmers
were shown in the current analysis. Tresno et al.’s (2012) study in university students
in Indonesia also found a comparable rate of self-harm across sexes, and a similar
rate to Gratz’s (2001) developmental study of the Deliberate Self-Harm Inventory
(DSHI) in the US. This indicates that self-harm may also be a widespread problem in
developing countries, with similar rates as in developed countries.
6.5.4 Cultural differences in aggressive behaviour
Despite the comparable effectiveness of mindfulness induction on direct aggression
following depletion, a different pattern of direct aggression was shown across
cultures. Specifically, compared to the British sample, the Indonesian sample
delivered significantly higher blast intensities under provocation and at baseline
level. Several mechanisms may have come into play.
In the TCRT task, the Indonesian participants might have simply complied with the
instruction to win the task, and were paying less attention to the disguised
provocation procedure (i.e., the noise blasts they received following each losing
trial). As a result, they might spontaneously deliver blast levels at any random point.
However, if this explanation holds, it is unclear why the standard deviation of the
Indonesians blast intensity was similar to the British (see Table 6.4).
More plausible is that the Indonesian may be less sensitive to the provocation
procedure in the TCRT task. This possibility may be related to the absence of
correlation between the Indonesian participants’ responses and levels of
provocation from the opponent (section 5.4.5.1, Chapter 5). Provocation has been
described as the major antecedent of aggression (Berkowitz, 1993; Geen, 2001), and
193
severe provocations typically generate more approval than unprovoked aggression
across cultures (Ramirez, 2007). Naturally, cultural differences would be smaller as
the level of provocations increased. While this would suggest that the Indonesians
may have lack of sensitivity to provocations, comparable scores were shown
between samples in trait SP. Thus the lack of provocation sensitivity of the
Indonesians seems to be specifically related to the TCRT method of provocation,
rather than to provocations in general. We did, however, found a typical sex
differences in both cultures, in which males delivered higher intensity blasts under
conditions of no and low provocation during the TCRT as opposed to females (see
Giancola & Parrott, 2008; Lawrence & Hutchinson, 2013a, 2013b).
We could not dismiss the second possibility that the Indonesian sample is, in fact,
more physically aggressive than the British, at least in the context of laboratory
aggression procedure that did not involve any apparent damage consequences. Due
to the increased fear of social evaluation when being observed acting out
aggressively, the Indonesian sample may only reduce aggression when seen. In the
TCRT task, aggressive behaviour is not directly observed. Relatedly, the Indonesians
might feel it is appropriate to chastise aggressive behaviour from the opponent in
the TCRT by delivering blasts as a punishment for aggressive behaviour. Higher blast
intensity of the Indonesians as opposed to the British was consistent when males
and females were analysed separately.
On the other hand, males in both cultures waited longer before delivering the
maximum blast as opposed to females. Giancola and Parrott (2008) posit that shock
intensity is a more explicit and blunt form of aggression, whereas shock duration is
more implicit, subtle, and less amenable to influences of social desirability and
impression management. The seemingly ambiguous nature of the latency measure is
arguably similar to shock duration, and thus perceived as more suitable measure for
female aggression than for male aggression across cultures due to its less overt
aspect.
194
The rating of indirect aggression for the Indonesians was equal to the British. This
runs counter the argument that conflict-avoidant values of collectivistic societies
may lead to greater indirect aggression (French et al., 2002), or the finding
elsewhere (Forbes et al., 2009) that decreases in individualism are associated with
decreases in both direct and indirect aggression. Examination on the specific indirect
aggression ratings revealed that even though Indonesians rated the opponent as
being more aggressive, they also rated the opponent as more fair and less
competitive. No significant culture or sex differences were found in rating of the
opponent’s skill or in the willingness to play with the same opponent again. The
inconclusive pattern of results in ratings of indirect aggression suggests that cultural
differences in aggression may be manifested into direct rather than indirect
aggression, at least in the current samples.
6.5.5 Cultural differences in self-control performance
Finally, we found that culture and participants’ sex played an important role on self-
control performance. Specifically, the Indonesians showed a better self-control
performance than the British prior to the TCRT task. This pattern (Indonesians
outperformed British), however, was particularly pronounced for males, whereas no
cultural differences were found in female’s ability to perform self-control task.
Gender role in self-control literature is commonly studied in a specific context of
delinquent and criminal behaviours (see e.g., Gibson, Ward, Wright, Beaver & Delisi,
2010). In contrast, ego-depletion studies have not typically investigated sex
differences in self-control performance (Hagger et al., 2010). In the current samples,
it is unlikely that sex differences in handgrip stamina were associated with trait self-
control (measured by Tangney et al.’s SCS, 2004), as reflected by the almost equal
self-control scores between sexes across cultures (M females = 3.13, SD = .47 vs. M
males = 3.06, SD = .55). Thus other factors may be more viable for this finding. For
instance, females were more able to maintain their baseline handgrip stamina than
males possibly because they are typically more compliant from an early age
195
(Kochanska, Coy & Murray, 2001), and our result implies that this tendency may be
similar across cultures.
Sex differences in self-control performance could also stemmed from a shift in
motivation (see Inzlicht & Schmeichel’s “process” model of depletion, 2012), or
other mediating variables that were not investigated in the current analysis, such as
subjective fatigue, perceived difficulty, or self-efficacy (Hagger et al., 2010). As
observed in both of our experiments, most participants considered the handgrip task
as a measure of physical strength. While furnishing baseline handgrip duration, male
participants tend to push themselves to hold the handgrip as long as they can, while
female participants tend to show less confident of their ability to perform the task.
On the subsequent measures of handgrip stamina, males showed more pressure to
compete with their own baseline duration, while females expressed more
confidence (as reflected in a typical comment, “This is much easier than what I
thought.”). Future work could fruitfully explore whether culture and sex differences
would emerge on other types of self-control task.
6.5.6 Strengths and limitations of the current analysis
The current analysis provided a statistical comparison of the strength of the benefit
of mindfulness induction in counteracting the effect of depletion on aggressive
behaviour and self-control performance across cultures. Cross-cultural comparison
of the personality measures in mindfulness, self-control, aggression, and self-harm
derived from Western operationalisation was also performed.
Apart from the argument that mindfulness should be more applicable in Indonesia
given its origin in Eastern contemplative traditions, our central theoretical construct
was the cultural differences in individualism vs. collectivism. We used the
individualism vs. collectivism distinction based on the characteristics of general
population, and did not measure these dimensions directly. Our sample consists of
university students, who are subjected to the influences of modernisation. For the
196
Indonesians, these influences may include a shift from traditional collectivism to
contemporary individualism, resulting in smaller cross-cultural differences than they
would be in the general population. Direct measures of cultural values should be
included in future cross-cultural study to test this possibility. Exploration of the role
of other constructs, such as religion, is also necessary, since religion may provide a
more comprehensive understanding on aggressive behaviour across cultures
(Ramirez, 2007).
Another limitation is that we did not operationalise the native British into their
culture groups (e.g., British Anglo-Saxons and British Asians). While our practice
appears to be common in cross-cultural work, intracultural differences are indeed
possible. Previous study (Thanzami & Archer, 2005) comparing British Anglo-Saxons
and British Asians from the Indian subcontinent suggested that those two samples
differed in their cultural orientation, although their belief about aggression was not
derived from a person’s cultural orientation. A future study could include measure of
beliefs about aggression in addition to direct measure of cultural values.
6.5.7 Conclusions
The current analyses revealed that mindfulness induction was effective at reducing
physical aggression and sustaining self-control performance under depleted self-
control resources across culture. Our findings also suggest that Eastern
conceptualisation of the constructs of mindfulness, self-control, aggression, and self-
harm may be related to the Western. It should be noted, however, that the
employment of the adapted TCRT task as a method of provocation might be less
effective in non-Western population, particularly in Indonesia. Even so, the TCRT task
appears to be effective as a means to demonstrate aggression that varies as a
function of sex, and is reduced via mindfulness interventions. The sensitivity of this
task in the Indonesians to pick up on differential provocation is less clear.
197
CHAPTER SEVEN
Summary and general discussion
In this chapter, we will reintroduce the original research questions, and then the
summaries of each study will be presented with particular focus as to how they
address the original research questions. The contribution of this thesis to the
literature on mindfulness, self-control, and aggression will be detailed, and the
limitations of the thesis and avenues for further research will be presented.
7.1. SUMMARY
7.1.1 Summary of each study
The primary question addressed by the thesis was: “What is the effect of
mindfulness on aggression?” Additionally, we investigated the potential role of self-
control on the potential link between mindfulness and aggression. In particular, the
thesis aimed to (i) examine the extent to which the benefits of mindfulness on
aggression may occur outside extensive training in mindfulness, (ii) directly compare
the benefits of mindfulness and self-control on aggression, (iii) assess the validation
of Western operationalisation of mindfulness, and self-control, on aggression, and
(iv) investigate any similarities between the mechanisms of aggression and self-harm
regarding mindfulness and self-control and individual differences.
In Study 1 (Chapter 2), we explored the relationship between self-reported measures
of mindfulness, self-control, and aggression (both to the self and to others). Data
from 241 participants (211 university students, 30 non-students) showed that
individuals who were more mindful and more self-controlled reported being less
aggressive typically. The influence that trait mindfulness had on general trait
aggression, as well as on physical aggression, anger, and hostility was mediated by
trait self-control. After accounting for self-control, mindfulness still predicted
significant variance in overall trait aggression, and anger, and hostility in particular,
198
but not in physical aggression. Mindfulness also decreased the likelihood of anger
amongst individuals who were more sensitive to provocations (SP), and of hostility
amongst those who were more sensitive to frustrations (SF). Additionally,
mindfulness and self-control reduced individuals’ tendency to harm themselves (self-
harm) in the same way as mindfulness and self control reduced aggression to others.
However, as in trait physical aggression, less self-harming behaviour occurred
primarily in those high in trait self-controlled regulation.
To examine the effect of mindfulness and self-control on aggressive behaviour, we
conducted three experimental studies. In these studies, aggressive behaviour was
provoked using the adapted Taylor Competitive Reaction Time (TCRT: Taylor, 1967)
task, in which participants were (i) exposed to no provocation, low provocation, and
high provocation trials from a bogus opponent, and (ii) given opportunity to deliver
or to not deliver a noise blast to the opponent in each winning trial. Participants’
aggressive behaviour was measured in terms of (i) blast intensity delivered to the
bogus opponent (as a main measure of direct physical aggression), (ii) maximum
blast latency (the number of trials participants waited before delivering the
maximum blast to the opponent, as an additional measure of direct aggression), and
(iii) indirect aggression (anonymous post TCRT reputation damage of the opponent).
In a validity study of the Taylor paradigm, Giancola and Parrott (2008) have
recommended shock or blast duration rather than blast intensity alone as a more
implicit measure of aggression. In addition, we also measured the impact of
mindfulness and self-control on self-control performance (i.e., handgrip task).
Study 2 (Chapter 3) employed a brief 10-min mindfulness walking exercise
(mindfulness induction vs. relaxation) and a 2-week training of improving posture
(self-control training vs. no training). Results from 99 (51 females) university
students showed that self-control training decreased blast intensity in response to
provocation. Reductions in provoked aggression were also predicted by lower levels
of trait SP. For female participants only, mindfulness induction predicted the delayed
delivery of maximum noise blasts to the opponent. These findings suggest that under
conditions of full self-control resources, self-control training increases people’s
capacity to resist from behaving aggressively (explicit aggression) when being
199
provoked. By contrast, mindfulness induction may be more reliable as a means to
reduce more subtle/implicit forms of aggression. While no complementary effect of
mindfulness and self-control exercises was found on aggressive behaviour, a
crossover interaction was shown on self-control performance (handgrip power).
Specifically, a better handgrip power was shown by those who received one
condition only (mindfulness induction or self-control training), but not by those who
received either both conditions or no condition. Interestingly, while trait aggression
and self-control were negatively related to self-harm in Study 1, Study 2 did not
show any relationship between either aggressive behaviour or self-control
performance and self-harm status.
In Study 2, all participants were operating under full self-control resource capacity.
Here, mindfulness had no clear influence on the intensity of aggression exhibited,
although the mindfulness indication has some influence on more implicit measures
of aggression (the delay duration before the maximum aggressive response was
delivered). In Study 3, the impact of mindfulness on resource-depleted participants
was examined. A different type of mindfulness induction was used (i.e., a 15-min of
mindfulness breathing exercise), and the results from 110 (58 females) native British
students revealed that mindfulness induction counteracted the effect of depletion
on blast intensity under no provocation and low provocation trials in the TCRT task,
but not under high provocations from the opponent. Individuals higher in trait SP
also delivered higher levels of blast intensity under low provocation trials. However,
contrary to our prediction, mindfulness induction increased the likelihood of
delivering the maximum blast earlier for depleted female participants. The positive
effect of mindfulness induction on self-control performance (assessed by changes in
handgrip duration relative to baseline) was independent of depletion condition.
Study 4 (Chapter 5) was a cross-cultural replication of Study 3. This study examined
the aggressive behaviour of 119 (60 females) Indonesian students recruited from a
university in Indonesia. Replicating the findings from Study 3, mindfulness induction
significantly moderated the link between ego-depletion and blast intensity on low
provocation trials. The moderation effect of mindfulness was also shown under high
provocation trials, but not in the absence of provocation. Interestingly, unlike in
200
Study 2 and 3, individual differences in SP did not predict blast intensity following
provocation. Mindfulness induction also decreased the likelihood of delivering the
maximum blast earlier for depleted Indonesian females. When measured before the
TCRT task, the positive effect of mindfulness induction on self-control performance
was shown in particular for depleted participants.
Data from the British and the Indonesian samples were statistically compared in
Chapter 6. The comparison showed that mindfulness induction reduced blast
intensities for depleted Indonesians to the same extent it affected their British
counterparts under all provocation trials. With the inclusion of participants’ sex,
mindfulness induction decreased the likelihood of delivering the maximum blast
earlier for depleted British participants compared to their Indonesian counterparts,
and for males as opposed to females. Thus in the context where individuals lack of
self-control resources, the role of mindfulness in reducing direct physical aggression
appears to be similar across cultures, whereas its effect in a subtle/implicit form of
aggression may be more evident for British participants, and for female than for
male aggression across cultures. Mindfulness and depletion conditions
independently influenced self-control performance (handgrip stamina) across
cultures. Additionally, the Indonesian sample delivered significantly higher blast
intensities across all provocation levels but also showed a better self-control
performance than the British sample, whereas no cultural differences were found on
the maximum blast latency, indirect aggression, or self-harm status.
7.1.2 Summary of findings
The thesis has examined the specific mechanisms through which mindfulness and
self-control may influence the production of aggressive/non-aggressive behaviour.
Trait mindfulness appears to play a unique role on reducing individual’s levels of
anger and hostility, whereas reductions in trait physical aggression and self-harm
may be more associated with self-control mechanisms (Study 1). Under depleted
self-control resource, mindfulness induction reduced direct physical aggression
(Studies 3 and 4). This effect was replicated across cultures (Chapter 6). Under full
201
resources of self-control, mindfulness induction only reduced the subtle/implicit
form of aggression (maximum blast latency for females), whereas self-control
training reduced direct aggression following provocation (Study 2). The effect
mindfulness induction on self-control performance (handgrip task), however, was
less predictable by the mechanism of self-control.
Neither mindfulness induction or self-control training (Study 2) nor mindfulness
induction or depletion (Study 3 and 4) influenced indirect aggression (anonymous
reputation damage to the opponent). Further, individual differences in SP predicted
provoked aggression (Study 2 and 3), except in the Indonesian samples (see Study 4).
Even though all studies reported that trait mindfulness and self-control were mostly
correlated to self-reported measures of aggression and self-harm, trait mindfulness
and self-control had no significant influence on the link between experimentally
activated mindfulness, self-control, and aggressive behaviour across the
experimental studies. The implications of these findings will be detailed in the
following section.
7.2. GENERAL DISCUSSION
7.2.1 Mindfulness, self-control, and aggression
As introduced in Chapter 1, mindfulness-based interventions have recently been
implemented for treatment of aggression in mental health and forensic settings (see
e.g., Howells, 2010; Howells et al., 2010; Shonin et al., in press). However, whether
and when the potential benefits of mindfulness can be separated from a general
mechanism of self-control are yet unclear, particularly in the context of extensive
training in mindfulness (Masicampo & Baumeister, 2007). Throughout this thesis, we
conducted a series of correlational and experimental studies to investigate the role
of mindfulness and self-control on aggression, based on the strength model of self-
control framework proposed by Baumeister and colleagues (1994, 2007). In this way,
it would be possible to examine whether the effect of mindfulness on aggression
202
may occur outside extensive training in mindfulness, while at the same time directly
compare the effect of mindfulness and self-control on aggression.
Resembling past research (e.g., Barnes et al., 2007; Borders et al., 2010; Brown &
Ryan, 2003; Kelly & Lambert, 2012), individual differences in mindfulness were
shown to be associated to lower propensity to aggress (Study 1). This association
was mediated by trait self-control, confirming the suggestion elsewhere (Borders et
al., 2010; Heppner et al., 2008) for the underlying mechanism of mindfulness in the
reductions of aggression. Importantly, we found that the presence of a mindful
quality uniquely reduced the experiences of anger and hostility, independent of
one’s dispositional ability in self-control. Even though the experience of anger is
typically short-lived (e.g., Averill, 1983; Tyson, 1998), it may produce a cognitive
residual of hostility (Buss & Perry, 1992). Moreover, ruminating on anger provoking
experiences may increase aggressive behaviour, even towards innocent others
(Bushman, 2002; Bushman et al., 2005). Our result implies that the capacity to be
aware of these experiences, with no direct attempt to control them, may decrease
the intensity of and increase the tolerance to these experiences, and thereby
reduces the preparedness to behave aggressively.
It should be noted, however, that presence of a mindful quality showed no
additional reduction in the overt component of aggression (i.e., trait physical
aggression) beyond the general ability to exert self-control (Study 1). Compared to
other components of the general trait aggression (i.e., verbal aggression, anger,
hostility: Buss & Perry, 1992), trait physical aggression has been demonstrated as the
strongest predictor of direct aggression in terms of shock intensities in the TCRT task
(Giancola & Parrott, 2008). Thus it is crucial to clarify whether the comparable
benefits of dispositional mindfulness and self-control on physical aggression would
be mirrored by experimentally activating levels of mindfulness and self-control.
The strength model of self-control has provided two two experimental pathways for
examining the role of self-control on aggressive behaviour, by (i) bolstering self-
control capacity (e.g., via self-control training and acute glucose consumption), and
(ii) weakening or depleting self-control capacity (Denson et al., 2012; DeWall et al.,
203
2011). In this thesis, their proposal was elaborated by including mindfulness.
Accordingly, we found that brief laboratory induction of state of mindfulness
reduced the production of direct physical aggression when individuals’ self-control
resource has been depleted (Studies 3 and 4). A similar effect was shown across
cultures (Chapter 6), indicating the efficacy of Western operationalisations of
mindfulness, self-control, and aggression in non-Western contexts.
In the absence of depletion (under full resources of self-control), the effect of state
mindfulness on aggressive behaviour was less salient (Study 2). That is, state
mindfulness only reduced the production of a subtle/implicit form of aggression (the
delay duration before the maximum aggressive response was delivered). In contrast,
training in self-control was effective in reducing direct physical aggression following
provocation (Study 2). This latter finding replicates the results from past studies
employing different types of self-control training, such as using nondominant hand in
mundane tasks (Denson et al., 2011) and verbal regulation of avoiding colloquialisms
(Finkel et al., 2009). Supporting the strength model proposal, the benefit of self-
control training on the reductions of aggression does not depend on the type of self-
control task being practiced, providing the practice requires the individuals to alter
their habitual responses (see e.g., Baumeister et al, 2006).
The notable divergence of the benefits of mindfulness induction under depleted and
full self-control resources requires further elaboration. Depletion may lead to an
impulsive, aggressive action in particular when the urge to aggress is high, such as
after being provoked (DeWall et al., 2011; Denson et al., 2012). In our studies,
depleted participants were shown to deliver higher blast intensities than non-
depleted ones after the experience of provocation across cultures (Study 3 and 4).
The cultivation of mindfulness serves a monitoring function to momentary (“in-the-
moment”) experience (e.g., Bishop et al., 2004; Brown et al., 2007a; Shapiro et al.,
2006). This may facilitate sustaining attention on the task at hand rather than
shifting motivation and attention from exerting control to acting on aggressive
impulses (see Inzlicht & Schmeichel’s “process” model of depletion, 2012). As
expected, the presence of state of mindfulness reliably counteracted the effect of
depletion on provoked aggression (Study 3 and 4). Under full self-control resource,
204
however, increases in self-control strength can be used effectively as a chief
resource to alter expressed aggression in response to provocation (Study 2).
It is also noteworthy that indirect aggression (anonymous reputation damage to the
opponent) was not influenced by mindfulness induction or self-control training
(Study 2), or by mindfulness induction or depletion (Study 3 and 4). Thus any
potential effect that mindfulness and self-control has on aggressive behaviour may
be less observable on indirect aggression.
Inconclusive pattern of results was obtained on the effect of mindfulness and self-
control on self-control performance (i.e., handgrip task). Mindfulness induction did
counteract the effect of depletion when self-control performance was measured
before the TCRT task (Study 4), but its positive effect after the TCRT was
independent of depletion condition (Study 3 and 4) and was even detrimental
amongst those who have performed training in self-control (Study 2). These findings
suggest that when individuals are required to perform a subsequent self-control act
after they have engaged in a provocation procedure, the complementary effect of
mindfulness and self-control is less marked, and may even reverse. Continuously
sustaining attention on the task at hand may be especially difficult for novice
meditator (Holzel et al., 2011; Sauer et al., 2013), and this effect seems to be more
pronounced after the experience of provocation.
Relatedly, the lack of participants’ formal experience with mindfulness meditation
might have contributed to absent of significant correlation between trait
mindfulness and self-control with experimentally activated mindfulness, self-control,
and aggressive behaviour across the experimental studies. As shown elsewhere (e.g.,
Carmody, Reed, Kristeller, & Merriam, 2008; Thompson & Waltz, 2007), at least
amongst individuals naive to meditation, state mindfulness during formal sitting
meditation was less related to their propensity to be mindfulness. A threshold level
of mindfulness may be required in order for novice meditators to show the ability to
be mindful during mindfulness induction.
205
7.2.2 Cultural differences in sensitivity to provocation and aggressive triggers
Cultures play a major role on the articulation of aggression and self-controlled acts.
Nevertheless, the benefit of mindfulness induction in attenuating the negative
impact of depletion on aggressive behaviour appears to be similar across cultures
(Chapter 6). The Indonesian sample (Study 4) reported very similar levels of
sensitivity to provocation (SP) to that within the British sample (Study 3). This
indicates that the Indonesian sample were no more sensitive to signals of
provocations than the British sample, despite hypothesised differences in the
tolerance of aggression as a means to facilitate the achievement of individual goals
(Forbes et al., 2009; Hofstede et al., 2010; Li et al., 2010).
According to Lawrence and colleagues, aggressive behaviour as a result of
provocation might be more accurately predicted by individual differences in SP than
by trait aggression alone (Lawrence & Hutchinson, 2013b; Lawrence & Hodgins,
2009). They also suggested that provocations and frustrations are not
interchangeable triggers of aggressive behaviour (Lawrence, 2006; Lawrence &
Hutchinson, 2013a). As predicted, we found that provoked aggression in the TCRT
task was affected by SP, such that individuals with higher levels of SP delivered
higher blast intensity to the opponent when they were being provoked, but not in
the absence of provocation (Study 2 and 3). Although individuals who reported
higher SP also reported higher levels of sensitivity to frustration (SF) and a greater
propensity to aggress, aggressive behaviour was neither predicted by SF nor by trait
aggression (Study 1 to 4). These findings provide further validation of the SP scale, in
addition to that shown by Lawrence and Hodgkins (2009) and Lawrence and
Hutchinson (2013b). In the Indonesian sample, however, the lack of significant
correlation between SP and provoked aggression may provide a challenge to the
applicability of the TCRT procedure in non-Western cultures.
With regards to the TCRT task, the Indonesian sample demonstrated the typical sex
differences associated with levels of provocation (i.e., males being more aggressive
than females under no provocation and low provocation, but not under high
provocation trials; see Bettencourt & Miller’s meta-analytic review, 2006). The usual
206
association between participants’ blast intensity and levels of provocation from the
opponent (e.g., Anderson et al., 2008; DeWall et al., 2010; Konrath et al., 2006;
Lawrence & Hutchinson, 2013a, 2013b; also our results in Study 2 and 3) was not
shown. Given that the validity of the Taylor paradigm as a measure of direct physical
aggression has been provided in Western countries (e.g., Giancola & Parrott, 2008),
it is plausible that the TCRT task might not influence aggression in the Indonesian
sample in the same way as has been shown in Western samples.
One way in which this difference in response to the TCRT task between British and
Indonesian samples might occur is due to the anticipation of social sanctions. As
pointed out by Lawrence and Hutchinson (2013b), the fact that the participants
never anticipate to meet their opponent in the TCRT task rarely occurs in real life. In
Evers et al.’s (2005) study, although anger provoking event (i.e., negative feedback
from a partner) increased the anger experience in male participants to the same
extent with in females, similarities in the anger expressions (i.e., the amount of hot
sauce allocated to the partner) were found only when participants did not expect to
meet the partner. Under a situation when they expected to meet the partner,
females expressed less anger than males. In the same way, the Indonesian people,
who commonly presumed that anger expression and direct aggression as a sign of
lack of self-control and inner strength (Koentjaraningrat, 1985; Magnis-Suseno,
1997), may only inhibit aggression expressions when the negative social implications
of these expressions are made salient. Relatedly, the implicit endorsement of the
participants’ aggressive responding in the TCRT task by an authority figure (i.e., the
experimenter; see Ritter & Elsea, 2005; Tedeschi & Quigley, 1996) might also be
particularly relevant for people from collectivist cultures.
Collectivist individuals should be more motivated to engage in self-control on daily
basis (Balliet & Joireman, 2010; Eisenberg & Zhou, 2000; Seeley & Gardner, 2003),
thus higher levels of self-control and less aggression should be expected. However,
repeated practice of self-control should be interspersed with rest (see Hagger et al.,
2010). If display of aggression is always frowned on, there is a possibility that people
might not develop the ability to moderate their aggression across situations as
much. Instead, their constant attempts to exert self-control over aggressive impulses
207
may produce a chronic state of ego-depletion. As recognised by Baumeister et al.
(2007b), “there are levels of depletion beyond which people may be unable to
control themselves effectively, regardless of what is at stake” (p.353). Altogether,
the lack of anticipated negative implications from the opponent, permission from
the authoritative figure, and a chronic state of depletion might have cause higher
display of aggressive behaviour in the Indonesian sample across all levels of
provocation from the opponent; and so decreasing the efficacy of the TCRT task.
7.2.3 Mindfulness, self-control, and self-harm
An additional aim of this thesis was to examine any similarities between the
mechanisms of aggression and self-harm, regarding mindfulness and self-control as
well as individual differences. A comparable rate of self-harm was shown across
cultures (Chapter 6). Individuals who self-harmed (having engaged in self-harming
behaviour more than five times: Gratz & Chapman, 2007) also reported greater
propensity to harm others (Studies 1, 2, and 4), supporting the positive association
between self-harm and aggression that has been outlined elsewhere using different
set of predictors (e.g., Barbui et al., 2009; Hillbrand, 2001; Placidi et al., 2001).
It has also been suggested that the risk assessment of self-harm as an emotional
regulation strategy should account for both self-control-related (i.e., ability to
engage in goal-directed behaviours and to inhibit impulsive behaviour) and
mindfulness-related components (i.e., the capacity to increase the awareness of and
acceptance of negative emotions: Gratz, 2007; Gratz & Roemer, 2004). Of the few
reported studies that investigated the role of both mindfulness and self-control
components in one sample, Slee, Spinhoven, et al. (2008) found a stronger mediating
role of self-control on the link between a brief Cognitive Behaviour Therapy (CBT)
and self-harm, but a much weaker mindfulness mediation. In this thesis, we also
found that while self-harming individuals tend to be less mindful and less self-
controlled (Study 1 and 2), declines in self-harming behaviour occurred primarily
through self-controlled regulation (Study 1).
208
In the experimental studies, we did not experimentally test self-harming behaviour.
Instead, we used individuals’ self-reported information regarding their propensity to
self-harm as a predictor for the role of mindfulness and self-control on aggressive
behaviour. Contrary to prediction, we found that self-harm status was neither
related to direct nor indirect aggression (Studies 2, 3, and 4). This implies that the
TCRT task may have not triggered sufficient emotional distress for self-harmers to
engage in aggressive responding. Self-harmers seem to posses more vulnerability to
frustrating or stressful events (Nock, 2009). These types of events may increase the
intensity and frequency of negative experiences. Individuals may then engage in self-
harming or aggressive behaviour to break the vicious cycle between these emotions
and rumination (Roberton et al., 2012; Selby et al., 2008, in press). Indeed, increased
rumination has been shown to predict increases in suicidality in a number of cross-
sectional, case-control and longitudinal studies (see Morrison & R. C. O’Connor’s
systematic review, 2008). It is unlikely that the TCRT task could engender intense
experiences which activate the link between rumination, self-harm, and aggression.
7.3 LIMITATIONS AND IMPLICATIONS FOR FUTURE RESEARCH
7.3.1 Sample characteristics
The experimental studies (Study 2, 3, and 4) in this thesis consisted of entirely
university students, thus limiting the generalisability of the findings. Results from
Study 1 (section 2.4.1, Chapter 2) indicated that non-students (12.40% of the total
participants) were higher in both trait mindfulness and self-control than student
participants (all significant at p < .0001). In the Mindfulness-Based Stress Reduction
(MBSR), individuals higher in trait mindfulness were shown to benefit more that
those with lower baseline levels (Shapiro, Brown, Thoresen, & Plante, 2011). A
similar effect might imply for those higher in trait self-control, as the complexity of
mindfulness interventions would require a high degree of self-control (for a review,
see Baer, 2003). On the other hand, a review on the benefits of laboratory-based
studies of mindfulness suggests that the beneficial effects of mindfulness may be
more noticeable amongst clinical populations than amongst healthy subjects (see
209
Keng et al., 2011). Either way, future work needs to examine whether a differential
effect of mindfulness induction on aggression would occur on diverse samples.
The experimental studies failed to recruit the optimum number of participants as
calculated by G*Power 3 (2013). With reference to the previous similar study (Friese
et al., 2012), however, the sample size in the cross-cultural studies (Study 3 and 4)
was sufficient. We also recruited an equal number of males and females in the
experimental studies to account for the possible sex differences that are prevalent in
studies using the TCRT tasks (see e.g., Anderson et al., 2008; Giancola & Parrott,
2008). To anticipate the issue with regards to this small sample size, a bootstrapping
procedure based on 5,000 resamples (A. F. Hayes, 2011, 2012a) was performed on
the data. Nevertheless, the effect of mindfulness induction on the link between
depletion and aggressive behaviour, although significant and consistent across
cultures, was considerably small in size (R2 increased due to the interaction was
around 3.29 to 3.79%, p < .05).
7.3.2 Brief laboratory mindfulness induction
Different types of mindfulness induction and control condition were employed in the
experimental studies. In Study 2, we used mindfulness walking exercise (Kabat-Zinn,
1994; Sujiva, 2000) and Progressive Muscle Relaxation (PMR: Jacobson, 1938) for the
control condition. In Study 3 and 4, we used mindfulness of breathing exercise
(Williams & Pennman, 2011) and a neutral educational information (e.g., Erisman &
Roemer, 2010). Thus it is plausible that the noticeable differential outcome across
these studies may be due to the lack of similarities of the mindfulness conditions.
As concluded in Keng et al.’s review (2011), a standard protocol for the mindfulness
instruction and control group in laboratory studies of mindfulness has not been
established; and only a limited number of these studies have explicitly assessed the
level of state mindfulness. The mindfulness of walking exercise is a novel technique
in laboratory settings, although it has been integrated in clinical training of
mindfulness (Centre for Mindfulness, 1995; Oxford Mindfulness Centre, 2002). The
210
relaxation exercise was given in Study 2 to limit the possibility that relaxation alone
could account for the impacts of mindfulness induction (Baer, 2003). A-1 month or
longer training in mindfulness was reported to produce additional effects over
relaxation training in terms of reducing distractive and ruminative thoughts and
behaviours (Jain et al., 2007) and reducing the emotional interference from
unpleasant stimuli (Ortner et al., 2007). In Study 2, no differences in participant’s
level of state mindfulness (TMS: Lau et al., 2006) were reported following a brief
period of mindfulness walking and relaxation exercises (i.e., 10 min). Moreover, both
exercises had no influence on participant’s blast intensity, although only mindfulness
induction predicted delayed delivery of the maximum blast. Exploring the point at
which the differential effect of mindfulness and relaxation could be observed, as well
as the comparability of various mindfulness exercises is indeed crucial.
In Study 3 and 4, we used one of the most common methods of laboratory
inductions of mindfulness (i.e., mindfulness of breathing; see e.g., Arch & Craske,
2006; Broderick, 2005; Feldman et al., 2010). As the relaxation procedure might
arguably be a very conservative control group, a neutral educational information was
given as control condition. A similar control condition has been employed in past
mindfulness, aggression, and ego-depletion studies. In Friese et al.’s (2012) ego-
depletion study with participants from a 3-day mindfulness meditation seminar, non-
mindfulness induction participants were asked to complete a neutral task (i.e., six
connect-the-dots figures forming mundane objects). In Heppner et al.’s (2008)
aggression study, control participants simply waited for approximately the same
time as the induction condition.
Future research could employ different types control condition, such as passive
progressive muscle relaxation (in which participants are instructed to relax muscle
groups without tensing) and loving-kindness meditation (a type of meditation that
focuses on feelings of warmth and care for oneself and others). One study
comparing three 15-min exercise in mindfulness of breathing, passive progressive
muscle relaxation, and loving-kindness meditation found that only mindfulness
meditation increased decentering (as assessed by Lau et al.’s TMS, 2006) and
reduced the association between frequency of repetitive thought and negative
211
reactions to thoughts (Feldman et al., 2010). Alternatively, various emotion
regulation strategies, such as cognitive reappraisal, distraction, unfocused attention,
or worry (see Keng et al., 2011) could also be employed. These latter conditions
were not employed in Study 3 and 4 because regulating emotions might cause
participants’ to spend their self-control resources (Baumeister, Vohs et al., 2007),
which may lead to an additional depleting effect.
7.3.3 Self-control task
A number of well-established self-control tasks were employed in this thesis. Self-
control performance was measured using a handgrip task (Study 2, 3, and 4; see
Hagger et al.’s meta-analysis, 2010). An attention control task was given as an ego-
depleting condition (Study 3 and 4; see DeWall et al., 2007; Finkel et al., 2009) and
an improving posture exercise was used as self-control training (Study 2; see e.g.,
Muraven et al., 1999). Despite the variety of these tasks, all imposed the necessity
for participants to exert self-control over their habitual responses, and were
reported to be successful in manipulating participants’ level of self-control in terms
of the extent to which participants have to control their attention (Study 3 and 4)
and adherence to the training (Study 2).
In Study 3 and 4, a 6-min period was chosen for the attention control task to induce
depleting effect. As noted by Hagger et al. (2010), the task duration in ego-depletion
studies typically lasted for less than 10 min, and the relationship between duration
of depleting task and impairment in self-control performance is only marginally
significant. Thus within the range of 10 min, spending a longer time on a depleting
task may not consume more of the self-control resource.
In Study 2, we aimed to compare the benefit of increased levels of mindfulness and
self-control on aggressive behaviour. Clearly we have to use longer than 10 min
duration in order to produce a positive effect of self-control exercise. However, we
have no a priori reference as to how long it would take for people to recover from
the depletion effect. Provided the limitation of self-control resource (Baumeister et
212
al., 1994; Baumeister, Vohs, et al., 2007), continuously practicing self-control without
sufficient amount of time to recover may be chronically depleting. Future work
should attempt to determine at what point practicing self-control task would stop
being ego-depleting and start becoming a strengthening of the self-control resource,
or even producing a chronic state of depletion.
On the other hand, practicing self-control acts for two weeks, presumably
interspersed with rest, appear to be sufficient to increase self-control strength (see
Baumeister et al., 2006). Therefore, in Study 2 we employed this standard duration
for self-control training, and compare its effect on aggressive behaviour to the effect
of a-10 min mindfulness induction. The incompatibility in time duration between
these treatments might have reduced the expected moderation effect of
mindfulness on the link between self-control training and aggression. Also, during
the 2-week interim period, non self-control training participants were not required
to perform any additional activities and were not contacted as much as the training
participants. Keeping an equal involvement between groups, as well as asking the no
training group to perform tasks that do not require self-control, such as solving math
problems or keeping a diary of the self-control acts they engage in (Muraven, 2010)
is noteworthy during the interim period.
In addition, our result also indicates the importance of self-controlled regulation on
the reductions of self-harm (Study 1). Future studies could fruitfully explore the
effect of self-control in daily activities (e.g., tracking eating habit, using nondominant
hand in a mundane task; see Baumeister et al., 2006) on self-harming behaviour.
7.3.4 Aggressive triggers
In this thesis, a loud noise blast was employed as a trigger for aggressive responding
due to the close resemblance between this trigger and physical provocation in real
life. We also provided three alternative ways to assess aggression in the TCRT task:
blast intensity, maximum blast latency, indirect aggression. The relevance of
alternative ways of assessing aggression other than blast intensity has been
213
suggested elsewhere (Giancola & Parrott, 2008). Yet, the efficacy of the TCRT task
appears to be limited in terms of its ability to (i) activate and differentiate between
levels of provoked aggression in the non-Western sample (Study 4), and (ii) activate
the link between individual differences in self-harm and aggressive behaviour.
Given the importance of social appraisal in collectivist societies (Hofstede et al.,
2010; Triandis, 1995), cross-cultural study employing the TCRT task could include
manipulation of participants’ expectations about whether or not they will meet the
opponent. Alternatively, providing participants with verbal insult from the
experimenter might also be particularly effective in collectivist samples. In this way,
the role of anticipated social sanctions for the expressions of aggression could be
activated. We did not include insult manipulation in the Indonesian sample for
comparability with previous findings in the literature and within the thesis.
Laboratory measures of self-harming behaviour, such as the Self-aggression
Paradigm (SAP) developed by Berman and colleagues (Berman & Walley, 2003;
McCloskey & Berman, 2003) could also be incorporated in future studies. The SAP is
based on the TCRT task, in which participants are provided the opportunity to self-
administer electric shocks each time they lose a trial. In that sense, self-harm may
also function as a mechanism for self-punishment (Gratz, 2003). A positive link has
been shown between shock intensity in the SAP and self-reported self-aggressive
ideation (Berman & Walley, 2003; Berman, Jones, & McCloskey, 2005) and clinician
ratings of suicidal and self-harming history (McCloskey & Berman, 2003). Replication
of the SAP and the TCRT task in a single study (providing option to blast both the
opponent and oneself) could help explain whether losing a competition would create
a sufficient stress for self-harmers to aggress against themselves and others.
Other validated laboratory procedure of aggression could also be applied, with the
aforementioned adjustments when tested across cultures and amongst self-harmers.
For instance, in the Point Subtraction Aggression Paradigm (Cherek, 1981; see also
Golomb, Cortez-Perez, Jaworski, Mednick, & Dimsdale, 2007), participants are asked
to play an economic game which involves pressing a particular button to earn a sum
of money or to deduct the same sum from an opponent. In the hot sauce paradigm
214
(Lieberman, Solomon, Greenberg, & McGregor, 1999), participants are required to
determine the amount of hot sauce to a provoking agent known to dislike spicy
foods. As in the Taylor paradigm, behaviour towards the partner is assumed to
reflect an immediate intent to cause harm through direct physical aggression.
7.3.5 Self-reported measures
All the self-reported measures used this thesis have been widely used validated,
particularly in Western countries (see section 2.3.3, Chapter 2). Cultural equivalency
in terms of sensitivity to provocations (SP), but not in sensitivity to frustrations (SF)
was shown in the Situational Triggers of Aggressive Responses (STAR: Lawrence,
2006) scale (Chapter 6). As SP did not predict aggressive behaviour in the TCRT task
for the Indonesian sample (Study 4), further studies may incorporate the STAR scale
to test whether this measure can reliably predict aggressive behaviour triggered by
different methods of provocation and frustration outside a Western context. A
similar suggestion also implies for other translated measures used in this thesis, the
Mindfulness Attention Awareness Scale (MAAS: Brown & Ryan, 2003), the Self-
Control Scale (SCS: Tangney et al., 2004), the Aggression Questionnaires (AQ: Buss &
Perry, 1992), the Deliberate Self-Harm Inventory (DSHI: Gratz, 2001), the Toronto
Mindfulness Scale (TMS: Lau et al., 2006), and the Positive Affect, Negative Affect
Schedule (PANAS: Watson et al., 1988), as we did not perform any additional testing
on the psychometric properties of these measures. The SCS in particular, appears to
include only self-controlled behaviour in serve of personal goals (see Appendix 2.1),
thus might be culturally bias when applied to collectivist individuals that are
expected to exert control as means of compliance with societal norms.
Information on participants’ self-harming behaviour in this thesis (as measured by
the DSHI) was limited to direct destruction of body tissue in the absence of suicidal
intent. The National Collaborating Centre for Mental Health (2012) has recently
suggested that the distinction of self-harming behaviour based on the
presence/absence of suicidal intent is unnecessary because the underlying
motivation of this behaviour is often unclear (see also Kapur et al., 2013). It is
215
plausible that a stronger pattern of association between self-harm and aggressive
behaviour would have been obtained if we include measures of self-harm in the
presence of suicidal intent. More severe symptoms of psychopathology and
psychosocial impairment have been shown by individuals with a history of both non-
suicidal self-harm and suicidal behaviour compared to those who engaged only in
self-harm in the absence of suicidal intent (see Hamza et al., 2012).
7.3.6 Future directions
The availability or temporary deficit in the self-control “resource” has been typically
measured indirectly, either from individuals’ post-depletion performance on self-
control tasks or on aggression tasks (see Hagger et al., 2010). As a result, the extent
to which mindfulness may replenish the self-control resource was also measured
indirectly throughout this thesis. Future studies should identify the underlying
emotional, cognitive, and biological processes that might come into play on the
association between mindfulness, self-control, and aggression.
According to the General Aggression Model (GAM: Anderson & Bushman, 2002),
aggressive behavior may be affected by genetic and other biological factors through
the influences of these factors on learning, decision-making, arousal, and affective
processes. For instance, testosterone in both males and females has been historically
linked to increases in physical aggression and mood, particularly following social
provocations (for a systematic review, see Archer, 1991; also Neave & D. B.
O'Connor’s recommendation, 2009, on experimentally elevated testosterone levels).
The link between testosterone and aggression was found to be mediated by reduced
activation in the brain regions associated with impulse control and self-regulation
(Mehta & Beer, 2010). Thus while this line of research has not been reported, it is
reasonable to expect that increases capacity to self-regulate due to mindfulness
interventions may influence the link between testosterone, self-control, and
aggression.
216
On the other hand, people’s plans, expectations, and goals may also influence
aggressive behaviour by influencing their appraisal and decision processes (Anderson
& Bushman, 2002). Thus one way through which aggression could be reduced is by
realising that there are (negative) potential consequences for behaving aggressively.
More expressed aggression was found amongst individuals who focus on the
immediate benefits and ignore the future consequences of their behaviour
(Bushman, Giancola, Parrot, & Roth, 2012; Joireman, Anderson, & Strathman, 2003).
These individuals also showed lower level of trait self-control (as measured by
Tangney et al.’s SCS, 2004) as well as greater susceptibility to ego-depletion effect
(Joireman, Balliet, Sprott, Spangenberg, & Shultz, 2008). As such, self-controlled
individuals may act less aggressively due to their strong concern about the future.
Conversely, mindfulness sustains people’s attention back to the present experience
rather than to the future or the past. Therefore, it is important to know what aspect
of the “here and now” mindfulness is focusing people on to reduce aggression.
Mindfulness may decrease aggression because it stops people from ruminating
towards aggression-triggering events in the past (Borders et al., 2010). However, it is
unclear whether keeping attention in the here and now would also mean stop
worrying about the negative consequences of aggression, which could increase
aggressive responses. Some reversed-items measure in the MAAS (Brown & Ryan,
2003; see appendix 2.1.) might appear to contradict people’s concern for the future
(e.g., “I get so focused on the goal I want to achieve that I lose touch with what I'm
doing right now to get there”). Measures of individual differences in rumination
(e.g., Ruminative Response Scale [RRS]: Nolen-Hoeksema & Morrow, 1990) and
anticipation of the future consequences (Consideration of Future Consequences
[CFC]: Strathman, Gleicher, Boninger, & Edwards, 1994) could be included to
anticipate where the interaction might be taking place.
Conversely, another way though which aggression could be reduced is by changing
one’s perspective about events in the past. Participants who were asked to reflect on
a provoking event from a “fly on the wall” (“self-distanced”) perspective, as opposed
to from a first-person (“self-immersed”) perspective reported fewer aggressive
thoughts and angry feelings, and displayed less aggressive behaviour (Mischkowski,
217
Kross, & Bushman, 2012). Self-distancing has also been proposed as a core element
that reduces the association between anger and rumination in Denson’s (2013)
multiple system model of angry rumination.
Self-distancing involves a shift in the content of people’s thought about their past, in
a sense that past experience is being reconstrued to promote insight and closure
rather than being recounted based on details of the experience (for a review, see
Kross & Ayduk, 2011). Arguably, self-distancing may qualify as a self-controlled
regulation because its aim is to change the self in order to avoid some undesirable
states (with regards to the past). It may also be related to mindfulness because both
involve a fundamental shift in one’s perspective (Kross, Gard, Deldin, Clifton, &
Ayduk, 2012). In mindfulness mode of awareness, however, no attempt is made to
change the content of consciousness based on self-reflection (Brown et al., 2007b;
Brown & Ryan, 2003). So any thinking that one does about the past is simply being
observed, without making direct effort to detach from the associated emotional
content. Future research could compare the benefit of practicing self-distancing,
mindfulness, and self-control on aggressive behaviour.
Concluding remarks
This thesis has been the first to replicate that the immediate benefits of mindfulness
on aggression can be measured outside extensive training in mindfulness. Only one
reported aggression study (Heppner et al., 2008) actively manipulated participants’
state of mindfulness. The thesis represents an important and original extension by (i)
directly comparing the effect of mindfulness and self-control training on aggressive
behaviour, (ii) investigating the role of mindfulness as one potential mechanism to
reduce aggression following self-control depletion, (iii) exploring whether the effect
of mindfulness on aggressive behaviour would differ for different levels of
provocation and different forms of aggression, (iii) including behavioural measure of
self-control performance as additional outcome, (iv) employing cross-cultural
samples, and (v) investigating any similarities between the mechanisms of aggression
and self-harm regarding mindfulness and self-control and individual differences.
218
At this point, it can be suggested that mindfulness induction is effective at reducing
physical aggression under ego-depleted condition, where individuals are “in
particular need for an efficient handling of mental resources” (Friese et al., 2012,
p.5). Since the conceptualisation of mindfulness, self-control, and aggression was
derived from Western operationalisations of these concepts, the findings also
suggest that the influences of these Western constructs may operate effectively in
non-Western settings. Information from this thesis may be useful for the
implementation of mindfulness and self-control for treatments of aggression and
self-harm across cultures.
219
REFERENCES
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology,53(1), 27–51. doi:10.1146/annurev.psych.53.100901.135231
Anderson, C. A., Buckley, K. E., & Carnagey, N. L. (2008). Creating your own hostile environment:A laboratory examination of trait aggressiveness and the violence escalation and the violenceescalation cycle. Personality and Social Psychology Bulletin, 34(4), 462-473.doi:10.1177/0146167207311282.
Arch, J. J., & Craske, M. G. (2006). Mechanisms of mindfulness: Emotion regulation following afocused breathing induction. Behaviour Research and Therapy, 44(12), 1849-1858.doi:10.1016/j.brat.2005.12.007
Archer, J. (1991). The influence of testosterone on human aggression. British Journal ofPsychology, 82(1), 1–28. doi: 10.1111/j.2044-8295.1991.tb02379.x
Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-analytic review.Review of General Psychology, 8(4), 291-322. doi:10.1037/1089-2680.8.4.291
Archer, J., & Coyne, S. M. (2005). An integrated review of indirect, relational, and socialaggression. Personality and Social Psychology Review, 9(3), 212–230.doi:10.1207/s15327957pspr0903_2
Archer, J., Kilpatrick, G., & Bramwell, R. (1995). Comparison of two aggression inventories.Aggressive Behavior, 21(5), 371–380. doi:10.1002/1098-2337(1995)21:5<371::AID-AB2480210506>3.0.CO;2-P
Averill, J. R. (1983). Studies on anger and aggression: Implications for theories of emotion.American Psychologist, 38(11), 1145−1160. doi:10.1037/0003-066X.38.11.1145
Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empiricalreview. Clinical Psychology: Science and Practice, 10(2), 125–143. doi:10.1093/clipsy.bpg015
Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of mindfulness by self-report: TheKentucky Inventory of Mindfulness Skills. Assessment, 11, 191-206.
Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self-reportassessment methods to explore facets of mindfulness. Assessment, 13(1), 27-45.doi:10.1177/1073191105283504
Baer, R. A., Smith, G. T., Lykins, E., Button, D., Krietemeyer, J., Sauer, S., … Williams, J.M. (2008).Construct validity of the five facet mindfulness questionnaire in meditating and nonmeditatingsamples. Assessment, 15(3), 329–342. doi:10.1177/1073191107313003
Balliet, D., & Joireman, J. (2010). Ego depletion reduces proselfs’ concern with the well-being ofothers. Group Processes Intergroup Relations, 13(2), 227–239. doi:10.1177/1368430209353634
Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44(9),1175-1184. doi:10.1037/0003-066X.44.9.1175
Barbui, C., Esposito, E., & Cipriani, A. (2009). Selective serotonin reuptake inhibitors and risk ofsuicide: A systematic review of observational studies. Canadian Medical Association Journal,180(3), 291–297. doi:10.1503/cmaj.081514
220
Barnes, S., Brown, K. W., Krusemark, E., Campbell, W. K., & Rogge, R. (2007). The role ofmindfulness in romantic relationship satisfaction and responses to relationship stress. Journal ofMarital and Family Therapy, 33(4), 482-500. doi:10.1111/j.1752-0606.2007.00033.x
Baron, B. A., & Richardson, D. R. (1994). Human aggression (2nd ed.). New York: Plenum Press.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in socialpsychological research: Conceptual, strategic, and statistical considerations. Journal ofPersonality and Social Psychology, 51(6), 1173-1182. doi:10.1037/0022-3514.51.6.1173
Bauer, I. M., & Baumeister, R. F. (2011). Self-regulatory strength. In K. D. Vohs & R. F.Baumeister (Eds.), Handbook of self-regulation: Research, theory, and applications, 2nd ed. (pp.64-82). London: The Guilford Press.
Baumeister, R. F. (1998). The self. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook ofsocial psychology, 4th ed. (pp. 680-740). New York: McGraw-Hill.
Baumeister, R. F. (2011). Self and identity: A brief overview of what they are, what they do, andhow they work. Annals of the New York Academy of Sciences, 1234, 48-55. doi:10.1111/j.1749-6632.2011.06224.x
Baumeister, R. F., & Alquist, J. L. (2009). Is there a downside to good self-control? Self andIdentity, 8(2), 115-130, doi:10.1080/15298860802501474
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the activeself a limited resource? Journal of Personality and Social Psychology, 74(5), 1252–1265.doi:10.1037/0022-3514.74.5.1252
Baumeister, R. F., Gailliot, M., DeWall, C. N., & Oaten, M. (2006). Self-regulation and personality:How interventions increase regulatory success, and how depletion moderates the effects oftraits on behavior. Journal of Personality, 74(6), 1773–1801. doi:10.1111/j.1467-6494.2006.00428.x
Baumeister, R. F., Heatherton, T. F., & Tice, D. M. (1994). Losing control: How and why peoplefail at self-regulation. San Diego, CA: Academic Press.
Baumeister, R. F., Schmeichel, B. J., & Vohs, K. D. (2007). Self-regulation and the executivefunction: The self as controlling agent. In A. W. Kruglanski and E. T. Higgins (Eds.), SocialPsychology: Handbook of Basic Principles, 2nd ed. (pp. 516–539). New York: Guilford.
Baumeister, R. F., & Vohs, K. D. (2007). Self-regulation, ego-depletion, and motivation. Socialand Personality Psychology Compass, 1(1), 115–128. doi:10.1111/j.1751-9004.2007.00001.x
Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control. CurrentDirections in Psychological Science, 16(6), 351-355. doi:10.1111/j.1467-8721.2007.00534.x
Beaton, D. E., Bombardier, C. Guillemin, F., & Ferraz, M. B. (2000). Guidelines for the process ofcross-cultural adaptation of self-report measures. Spine, 25(24), 3186–3191. Retrieved fromhttp://journals.lww.com/spinejournal/pages/default.aspx
Bergeron, N., & Schneider, B. H. (2005). Explaining cross-national differences in peer-directedaggression: A quantitative synthesis. Aggressive Behavior, 31(2), 116–137.doi:10.1002/ab.20049
221
Bergomi, C., Tschacher, W., & Kupper, Z. (2012). The assessment of mindfulness with self-reportmeasures: Existing scales and open issues. Mindfulness. Advance online publication.doi:10.1007/s12671-012-0110-9
Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and reformulation.Psychological Bulletin, 106(1), 59–73. doi:10.1037/0033-2909.106.1.59
Berkowitz, L. (1990). On the formation and regulation of anger and aggression: A cognitive-neoassociationistic analysis. American Psychologist, 45(4), 494–503, doi:10.1037/0003-066X.45.4.494.
Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. New York: McGraw-Hill.
Berkowitz, L. (2008). On the consideration of automatic as well as controlled psychologicalprocesses in aggression. Aggressive Behavior, 34(2), 117–129. doi:10.1002/ab.20244.
Berman, M. E., & Walley, J. C. (2003). Imitation of self-aggressive behavior: An experimental testof the contagion hypothesis. Journal of Applied Social Psychology, 33(5), 1036–1057.doi:10.1111/j.1559-1816.2003.tb01937.x
Berman, M., Jones, G., & McCloskey, M. (2005). The effects of diazepam in self-aggressivebehavior. Psychopharmacology, 178(1), 100-106. doi:10.1007/s00213-004-1966-8
Bettencourt, B. A., & Miller, N. (1996). Gender differences in aggression as a function ofprovocation: A meta-analysis. Psychological Bulletin, 119(3), 422-447. doi:10.1037/0033-2909.119.3.422
Bettencourt, B. A., Talley, A., Benjamin, A. J., & Valentine, J. (2006). Personality and aggressivebehavior under provoking and neutral conditions: A meta-analytic review. Psychological Bulletin,132(5), 751-777. doi:10.1037/0033-2909.132.5.751
Bien, T., & Didonna, F. (2009). Appendix A: Mindfulness practice. In F. Didonna (Ed.), ClinicalHandbook of Mindfulness (pp. 477-488). New York: Springer.
Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., ... Devins, G. (2004).Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11(3),230–241. doi:10.1093/clipsy/bph077
Bjelland, I., Dahl, A. A., Haug, T. T., & Necklemann, D. (2002). The validity of the Hospital Anxietyand Depression Scale: An updated literature review. Journal of Psychosomatic Research, 52(2),549–556. doi:10.1016/S0022-3999(01)00296-3
Björkqvist K., (1994). Sex differences in physical, verbal and indirect aggression: A review ofrecent research. Sex Roles, 30(3-4), 177-188. doi:10.1007/BF01420988
Björkqvist, K., Lagerspetz, K. M., & Kaukianen, A. (1992). Do girls manipulate and boys fight?Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18(2),117-127. doi:10.1002/1098-2337(1992)18:2<117::AID-AB2480180205>3.0.CO;2-3
Black, D.S., Sussman, S., Johnson, C.A., & Milam, J. (2012) Psychometric assessment of theMindful Attention Awareness Scale (MAAS) among Chinese adolescents. Assessment, 19(1), 42-52. doi:10.1177/1073191111415365
222
Borders, A., Earleywine, M., & Jajodia, A. (2010). Could mindfulness decrease anger, hostility,and aggression by decreasing rumination? Aggressive Behavior, 36(1), 28–44.doi:10.1002/ab.20327
Bowen, S., Chawla, N., & Marlatt, G. A. (2010). Mindfulness-Based Relapse Prevention foraddictive behaviors: A clinician’s guide. New York: Guilford Press.
Bowlin, S.L., & Baer, R.A. (2012). Relationships between mindfulness, self-control, andpsychological functioning. Personality and Individual Differences, 52(3), 411–415.doi:10.1016/j.paid.2011.10.050
Brittlebank, A. D., Cole, A., Hassanyeh, F., Kenny, M., Simpson, D., Scott, J. (1990). Hostility,hopelessness and deliberate self-harm: A prospective follow-up study. Acta PsychiatricaScandinavica, 81(3), 280–283. doi:10.1111/j.1600-0447.1990.tb06497.x
Broderick, P.C. (2005). Mindfulness and coping with dysphoric mood: Contrasts with ruminationand distraction. Cognitive Therapy and Research, 29(5), 501–510. doi:doi:10.1007/s10608-005-3888-0
Brown, K. W., & Cordon, S. (2009). The phenomenological nature and emotional correlates ofmindfulness. In F. Didonna (Ed.), Clinical Handbook of Mindfulness (pp. 59-81). New York:Springer.
Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: Mindfulness and its role inpsychological well-being. Journal of Personality and Social Psychology, 84(4), 822–848.doi:10.1037/0022-3514.84.4.822
Brown, K. W., & Ryan, R. M. (2004). Perils and promise in defining and measuring mindfulness:Observations from experience. Clinical Psychology: Science and Practice, 11(3), 242–248.doi:10.1093/clipsy.bph078
Brown, K. W., Kasser, T., Linley, P. A., & Ryan, R. M. (2009). When what one has is enough:Mindfulness, desire discrepancies, and subjective well-being. Journal of Research in Personality,43(5), 727–736. doi:10.1016/j.jrp.2009.07.002
Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007a). Addressing fundamental questions aboutmindfulness. Psychological Inquiry, 18(4), 272–281. doi:10.1080/10478400701703344
Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007b). Mindfulness: Theoretical foundations andevidence for its salutary effects. Psychological Inquiry, 18(4), 211–237.doi:10.1080/10478400701598298
Brown, K. W., Ryan, R. M., Loverich, T. M., Biegel, G. M., & West, A. M. (2011). Out of thearmchair and into the streets: Measuring mindfulness advances knowledge and improvesinterventions: Reply to Grossman (2011). Psychological Assessment, 23(4), 1041-1046.doi:10.1037/a0025781
Brown, M. Z., Comtois, K. A., & Linehan, M. M. (2002). Reasons for suicide attempts andnonsuicidal self-injury in women with borderline personality disorder. Journal of AbnormalPsychology, 111(1), 198–202.
Burton, V. S., Cullen, F. T., Evans, T. D., Alarid, L. F., & Dunaway, R. G. (1998). Gender, self-control, and crime. Journal of Research in Crime and Delinquency, 35(2), 123-147.doi:10.1177/0022427898035002001
223
Bushman, B. J. (2002). Does venting anger feed or extinguish the flame? Catharsis, rumination,distraction, anger, and aggressive responding. Personality and Social Psychology Bulletin, 28(6),724–731. doi:10.1177/0146167202289002
Bushman, B. J., & Anderson, C. A. (2001). Is it time to pull the plug on the hostile versusinstrumental aggression dichotomy? Psychological Review, 108(1), 273–79. doi:10.1037/0033-295X.108.1.273
Bushman, B. J., Baumeister, R. F., & Phillips, C. M. (2001). Do people aggress to improve theirmood? Catharsis beliefs, affect regulation opportunity and aggressive responding. Journal ofPersonality and Social Psychology, 81(1), 17-32. doi:10.1037/0022-3514.81.1.17
Bushman, B. J., Bonacci, A. M., Pederson, W. C., Vasquez, E. A., & Miller, N. (2005). Chewing on itcan chew you up: Effects of rumination on triggered displaced aggression. Journal of Personalityand Social Psychology, 88(6), 969–983. doi:10.1037/0022-3514.88.6.969
Bushman, B. J., Giancola, P. R., Parrot, D. J., Roth, R. M. (2012). Failure to consider futureconsequences increases the effects of alcohol on aggression. Journal of Experimental SocialPsychology, 48(2), 591–595. doi:10.1016/j.jesp.2011.11.013
Buss, A. H., & Warren, W. L. (2000). The aggression questionnaire manual. Los Angeles: WesternPsychological Services.
Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality and SocialPsychology, 63(3), 452–459. doi:10.1037/0022-3514.63.3.452
Cairns, R. B., Cairns, B., Neckerman, H., Ferguson, L., & Gariepy, J. (1989). Growth andaggression: I. Childhood to early adolescence. Developmental Psychology, 25(2), 320-330.doi:10.1037/0012-1649.25.2.320
Card, N. A., Stucky, B. D., Sawalani, G. M., & Little, T. D. (2008.) Direct and indirect aggressionduring childhood and adolescence: A meta-analytic review of gender differences,intercorrelations, and relations to maladjustment. Child Development, 79(5), 1185 – 1229.doi:10.1111/j.1467-8624.2008.01184.x
Cardaciotto, L. A., Herbert, J. D., Forman, E. M., Moitra, E., & Farrow, V. (2008). The assessmentof present-moment awareness and acceptance: The Philadelphia mindfulness scale. Assessment,15(2), 204. doi:10.1177/1073191107311467
Carmody, J., & Baer, R. A. (2009). How long does a mindfulness-based stress reduction programneed to be? A review of class contact hours and effect sizes for psychological distress. Journal ofClinical Psychology, 65(6), 627–638. doi:10.1002/jclp.20555
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework forpersonality-social, clinical and health psychology. Psychological Bulletin, 92(1), 111–135.doi:10.1037/0033-2909.92.1.111
Carver, C. S., & Scheier, M. F. (2011). Self-regulation of action and affect. In K. D. Vohs & R. F.Baumeister (Eds.), Handbook of selfregulation (pp. 3–21). New York: Guilford.
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood toadulthood. Journal of Personality and Social Psychology, 78(1), 158–172. doi:10.1037/0022-3514.78.1.158
224
Centre for Mindfulness in Medicine, Health Care and Society (University of MassachusettsMedical School). (1995). Stress Reduction Program. Retrieved from http://www.umassmed.edu/cfm/home/index.aspx
Chambers, R., Gullone, E., & Allen, N. B. (2009). Mindful emotion regulation: An integrativereview. Clinical Psychology Review, 29(6), 560–572. doi:10.1016/j.cpr.2009.06.005
Chapman, A. L. & Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences ofdeliberate self-harm and suicide attempts. Suicide Life Threat Behavior, 37(5), 543-52.doi:10.1521/suli.2007.37.5.543
Chapman, A. L., Gratz, K. L., & Brown, M. Z. (2006). Solving the puzzle of deliberate self-harm:The experiential avoidance model. Behaviour Research and Therapy, 44(3), 371-394.doi:10.1016/j.brat.2005.03.005
Cherek D. R. (1981). Effects of smoking different doses of nicotine on human aggressivebehavior. Psychopharmacology, 75(4), 339–345. doi:10.1007/BF00435849
Chiesa, A., & Malinowski, P. (2011). Mindfulness-based approaches: Are they all the same?Journal of Clinical Psychology, 67(4), 1–21. doi:10.1002/jclp.20776
Chiesa, A., Serretti, A., & Jakobsen, J.C. (2013). Mindfulness: Top–down or bottom–up emotionregulation strategy? Clinical Psychology Review, 33(1), 82–96. doi:10.1016/j.cpr.2012.10.006
Chilvers, J., Thomas, C., & Stanbury, A. (2011). The impact of a ward-based mindfulnessprogramme on recorded aggression in a medium secure facility for women with learningdisabilities. Journal of Learning Disabilities and Offending Behaviour, 2(1), 27–41.doi:10.5042/jldob.2011.0026
Christopher, M. S., Charoensuk, S., Gilbert, B. D., Neary, T. J., & Pearce, K. L. (2009). Mindfulnessin Thailand and the United States: A case of apples versus oranges? Journal of ClinicalPsychology, 65(6), 590–612. doi:10.1002/jclp.20580
Cohn, A. M., Jakupcak, M., Seibert, L. A., Hildebrandt, T. B., & Zeichner, A. (2010). The role ofemotion dysregulation in the association between men's restrictive emotionality and use ofphysical aggression. Psychology of Men and Masculinity, 11(1), 53–64. doi:10.1037/a0018090
Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessmentusing the Revised NEO Personality Inventory. Journal of Personality Assessment, 64(1), 21–50.doi:10.1207/s15327752jpa6401_2
Creswell, J. D., Way, B. M., Eisenberger, N. I., & Lieberman, M. D. (2007). Neural correlates ofdispositional mindfulness during affect labeling. Psychosomatic Medicine, 69(6), 560-565.doi:10.1097/PSY.0b013e3180f6171f
Crick, N. R., & Grotpeter, J. (1995). Relational aggression, gender, and social-psychologicaladjustment. Child Development, 66(3), 710-722. doi:10.1111/j.1467-8624.1995.tb00900.x
Davis, D. M., & Hayes, J. A. (2011). What are the benefits of mindfulness? A practice review ofpsychotherapy-related research. Psychotherapy, 48(2), 198–208. doi:10.1037/a0022062
De Ridder, D. T., Lensvelt-Mulders, G., Finkenauer, C., Stok, F. M., & Baumeister, R. F. (2012). Ameta-analysis of how trait self-control relates to a wide range of behaviors. Personality andSocial Psychology Review, 16(1), 76-99. doi:10.1177/1088868311418749
225
Denson, T. F. (2013). The multiple systems model of angry rumination. Personality and SocialPsychology Review, 17(2), 103-123. doi:10.1177/1088868312467086
Denson, T. F., Capper, M. M., Oaten, M., Friese, M., & Schofield, T. P. (2011). Self-controltraining decreases aggression in response to provocation in aggressive individuals. Journal ofResearch in Personality, 45(2), 252–256. doi:10.1016/j.jrp.2011.02.001
Denson, T. F., DeWall, C. N., & Finkel, E. J. (2012). Self-control and aggression. Current Directionsin Psychological Science, 21(1), 20–25. doi:10.1177/1088868312467086
Denson, T. F., von Hippel, W., Kemp, R. I., & Teo, L. S. (2010). Glucose consumption decreasesimpulsive aggression in response to provocation in aggressive individuals. Journal ofExperimental Social Psychology, 46(6), 1023–1028. doi:10.1016/j.jesp.2010.05.023
DeWall, C. N., Buckner, J. D., Lambert, N. M., Cohen, A. S., & Fincham, F. D. (2010). Bracing forthe worst, but behaving the best: Social anxiety, hostility, and behavioral aggression. Journal ofAnxiety Disorders, 24(2), 260–268. doi:10.1016/j.janxdis.2009.12.002
DeWall, C. N., Baumeister, R. F., Stillman, T., & Gailliot, M. T. (2007). Violence restrained: Effectsof self-regulation and its depletion on aggression. Journal of Experimental Social Psychology,43(1), 62-76. doi:10.1016/j.jesp.2005.12.005
DeWall, C. N., Finkel, E. J. & Denson, T. F. (2011). Self-control inhibits aggression. Social andPersonality Psychology Compass, 5(7), 458–472. doi:10.1111/j.1751-9004.2011.00363.x
Dryden, W., & Still, A. (2006). Historical aspects of mindfulness and self-acceptance inpsychotherapy. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 24(1), 3-28.doi:10.1007/s10942-006-0026-1
Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the convergent validity of self-controlmeasures. Journal of Research in Personality, 45(3), 259-268. doi:10.1016/j.jrp.2011.02.004
Eagly, A. H., & Wood, W. (2009). Sexual selection does not provide an adequate theory of sexdifferences in aggression. Behavioral and Brain Sciences, 32(3–4), 276–277.doi:10.1017/S0140525X09990264
Erisman, S. M., & Roemer, L. (2010). A preliminary investigation of the effects of experimentallyinduced mindfulness on emotional responding to film clips. Emotion, 10(1), 72–82. doi:10.1037/a0017162
Evers, C., Fischer, A. H., Rodriguez Mosquera, P. M., & Manstead, A. S. R. (2005). Anger andsocial appraisal. A spicy sex difference? Emotion, 5(3), 258–266. doi:10.1037/1528-3542.5.3.258
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.G. (2009). Statistical power analyses using G*Power3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149-60.doi:10.3758/BRM.41.4.1149
Favazza, A. R. (1998). The coming of age of self-mutilation. Journal of Nervous and MentalDiseases, 186(5), 259–268. 10.1097/00005053-199805000-00001
Favazza, A. R. (2012, March). Nonsuicidal self-injury: How categorization guides treatment.Current Psychiatry, 11(3). Retrieved fromhttp://currentpsychiatry.com/pdf/1103/1103CP_Favazza.pdf
226
Feldman G., Greeson J., & Senville J. (2010). Differential effects of mindful breathing,progressive muscle relaxation, and loving-kindness meditation on decentering and negativereactions to repetitive thoughts. Behaviour Research and Therapy, 48(10), 1002-1011.doi:10.1016/j.brat.2010.06.006
Fetterman, A. K., Robinson, M. D., Ode, S., & Gordon, K. H. (2010). Neuroticism as a risk factorfor behavioral dysregulation: A mindfulness-mediation perspective. Journal of Social and ClinicalPsychology, 29(3), 301-321. doi:10.1521/jscp.2010.29.3.301
Finkel, E. J., DeWall, C. N., Slotter, E. B., Oaten, M., & Foshee, V. A. (2009). Self-regulatory failureand intimate partner violence perpetration. Journal of Personality and Social Psychology, 97(3),483–499. doi:10.1037/a0015433
Fix, R. L., & Fix, S. T. (2013). The effects of mindfulness-based treatments for aggression: Acritical review. Aggression and Violent Behavior, 18(2), 219–227. doi:10.1016/j.avb.2012.11.009
Fliege, H., Kocalevent, R., Walter, O. B., Beck, S., Gratz, K. L., Gutierrez, P., & Klapp, B. F. (2006).Three assessment tools for deliberate self-harm and suicide behavior: Evaluation andpsychopathological correlates. Journal of Psychosomatic Research, 61(1), 113–121.doi:10.1016/j.jpsychores.2005.10.006
Forbes, G., Zhang, X., Doroszewicz, K., & Haas, K. (2009). Relationships between individualism-collectivism, gender, and direct or indirect aggression: A study in China, Poland, and the US.Aggressive Behavior, 35(1), 24-30. doi:10.1002/ab.20292
French, D. C., Jansen, E. A., & Pidada, S. (2002). United States and Indonesian children’s andadolescents’ reports of relational aggression by disliked peers. Child Development, 73(4), 1143–1150. doi:10.1111/1467-8624.00463
Friese, M., Messner, C., & Schaffner, Y. (2012). Mindfulness meditation counteracts self-controldepletion. Consciousness and Cognition, 21(2), 1016–1022. doi:10.1016/j.concog.2012.01.00
G*Power 3 (Version 3.1.7) (Software). (2013) Department of Experimental Psychology Heinrich-Heine-University, Düsseldorf: Germany. Retrieved from http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3/download-and-register
Gailliot, M. T., Plant, E. A., Butz, D. A., & Baumeister, R. F. (2007). Increasing self-regulatorystrength can reduce the depleting effect of suppressing stereotypes. Personality and SocialPsychology Bulletin, 33(2), 281–294. doi:10.1177/0146167206296101
Geen, R. G. (2001). Human aggression (2nd ed.). Buckingham, England: Open University Press.
Gerevich, J., Bácskai, E., & Czobor, P. (2007). The generalizability of the Buss–Perry AggressionQuestionnaire. International Journal of Methods in Psychiatric Research, 16(3), 124–136.doi:10.1002/mpr.221
Giancola, P. R. (2000). Executive functioning: A conceptual framework for alcohol-relatedaggression. Experimental and Clinical Psychopharmacology, 8(4), 576-59, doi:10.1037/1064-1297.8.4.576
Giancola, P. R., & Parrott, D.J. (2008). Further evidence for the validity of the Taylor aggressionparadigm. Aggressive Behavior, 34(2), 214-229. doi:10.1002/ab.20235
227
Giancola, P. R., & Zeichner, A. (1995). Construct validity of a competitive reaction‐timeaggression paradigm. Aggressive Behavior, 21(3), 199-204. doi:10.1002/1098-2337(1995)21:3<199::AID-AB2480210303>3.0.CO;2-Q
Gibson, C. L., Ward, T. J., Wright, J. P., Beaver, K. M., & Delisi, M. (2010). Where does gender fitin the measurement of self-control? Criminal Justice and Behavior, 37(8), 883-903.doi:10.1177/0093854810369082
Gjersing, L., Caplehorn, J. R. M., & Clausen, T. (2010). Cross-cultural adaptation of researchinstruments: Language, setting, time and statistical considerations. BMC Medical ResearchMethodology, 10, 13. doi:10.1186/1471-2288-10-13.
Golomb B. A., Cortez-Perez, M., Jaworski, B. A, Mednick, S., & Dimsdale, J. (2007). Pointsubtraction aggression paradigm: Validity of a brief schedule of use. Violence and Victims, 22(1),95-103. doi:10.1891/vv-v22i1a00
Goodwin, C.J. (2010). Research in psychology: Methods and design (6th ed.). New York: JohnWiley and Sons.
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-based studies?A comparative analysis of six preconceptions about internet questionnaires. AmericanPsychologist, 59(2), 93-104. doi:10.1037/0003-066X.59.2.93
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: StanfordUniversity Press.
Grasmick, H. G., Tittle, C. R., Bursik, R. J. J., & Arneklev, B. J. (1993). Testing the core empiricalimplications of Gottfredson and Hirschi’s general theory of crime. Journal of Research in Crimeand Delinquency, 30(1), 5-29. doi:10.1177/0022427893030001002
Gratz, K. L. (2001). Measurement of deliberate self-harm: Preliminary data on the DeliberateSelf-harm Inventory. Journal of Psychopathology and Behavioral Assessment, 23(4), 253–263.doi:10.1023/A:1012779403943
Gratz, K. L. (2003). Risk factors for and functions of deliberate self-harm: An empirical andconceptual review. Clinical Psychology: Science and Practice, 10(2), 192–205.doi:10.1093/clipsy.bpg022
Gratz, K. L. (2007). Targeting emotion dysregulation in the treatement of self-injury. Journal ofClinical Psychology in Session, 63(11), 1091–1103. doi:10.1002/jclp.20417
Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment of emotion regulation anddysregulation: Development, factor structure, and initial validation of the Difficulties in EmotionRegulation Scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 41–54.doi:10.1023/B:JOBA.0000007455.08539.94
Gross, J. J. (2002). Emotion regulation: affective, cognitive, and social consequences.Psychophysiology, 39(3), 281-291. doi:10.1017/S0048577201393198
Hagger, M.S., Wood, C., Chris, S., & Chatzisarantis, N.L.D. (2010). Ego-depletion and the strengthmodel of self-control: A meta-analysis. Psychological Bulletin. 136(4), 495-525.doi:10.1037/a0019486
228
Hamza C. A., Stewart, S. L, Willoughby, T. (2012). Examining the link between nonsuicidal self-injury and suicidal behavior: A review of the literature and an integrated model. ClinicalPsychology Review, 32(6), 482–495. doi:10.1016/j.cpr.2012.05.003.
Hawton, K., Rodham, K., Evans, E., & Weatherall, R. (2002). Deliberate self-harm in adolescents:Self-report survey in schools in England. British Medical Journal, 325, 1207–1211.doi:10.1136/bmj.325.7374.1207
Hawton, K., Saunders, K. E. A., & O’Connor, R.C. (2012). Self-harm and suicide in adolescents.The Lancet, 379, 2373–2382. doi:10.1016/S0140-6736(12)60322-5
Hayes, A. F. (2012a). PROCESS for SPSS and SAS [Macro]. Retrieved fromhttp://www.afhayes.com/introduction-to-mediation-moderation-and-conditional-process-analysis.html
Hayes, A. F. (2012b). PROCESS: A versatile computational tool for observed variable mediation,moderation, and conditional process modeling [White paper]. Retrieved fromhttp://www.afhayes.com/public/process2012.pdf
Hayes, A. M., & Feldman, G. (2004). Clarifying the construct of mindfulness in the context ofemotion regulation and the process of change in therapy. Clinical Psychology: Science andPractice, 11(3), 255–262.
Hayes, S. C. (2002). Acceptance, mindfulness, and science. Clinical Psychology: Science andPractice, 9(1), 101-106. doi:10.1093/clipsy.9.1.101.doi:10.1093/clipsy.bph080
Hayes, S. C. (2004). Acceptance and Commitment Therapy and the new behavior therapies:Mindfulness, acceptance, and relationship. In S. C. Hayes, V. M. Follette, & M. M. Linehan (Eds.),Mindfulness and acceptance: Expanding the cognitive-behavioral tradition. New York: Guilford.
Hayes, S. C., Follette, V. M., & Linehan, M. M. (Eds.). (2004). Mindfulness and acceptance:Expanding the cognitive-behavioral tradition. New York: Guilford.
Hayes, S. C., Strosahl, K., & Wilson, K. G. (2011). Acceptance and Commitment Therapy: Theprocess and practice of mindful change (2nd ed.). New York: Guilford Press.
Heatherton, T. F., & Wagner, D. D. (2011). Cognitive neuroscience of self-regulation failure.Trends in Cognitive Sciences, 15(3), 132–139.
Heppner, W. L., Kernis, M. H., Lakey C. E., Campbell, W. K., Goldman, B. M., Davis P. J, & Cascio,E. V. (2008). Mindfulness as a means of reducing aggressive behavior: Dispositional andsituational evidence. Aggressive Behavior, 34(5), 486–496. doi:10.1002/ab.20258
Hill, K. & Dallos, R. (2012). Young people’s stories of self–harm: A narrative study. Clinical ChildPsychology and Psychiatry, 17(3), 459-475. doi:10.1177/1359104511423364
Hillbrand, M. (2001). Homicide-suicide and other forms of co-occurring aggression against selfand against others. Professional Psychology: Research and Practice, 32(6), 626-635.doi:10.1037/0735-7028.32.6.626
Hofstede G. (1980.) Culture’s consequences. California: Sage.
Hofstede G. (2010). Cultural dimensions. Retrieved from http://geert-hofstede.com/countries.html
229
Hofstede, G., Hofstede, G. J., & Minkov, M. (2010). Cultures and organizations: Software of themind (3rd ed.). New York: McGraw-Hill.
Holzel, B. K., Lazar, S. W., Gard, T., Schuman-Olivier, Z., Vago, D. R., & Ott, U. (2011). How doesmindfulness meditation work? Proposing mechanisms of action from a conceptual and neuralperspective. Perspectives on Psychological Science, 6(6), 537–59.doi:10.1177/1745691611419671
Howells, K. (2010). The ‘third wave’ of cognitive-behavioural therapy and forensic practice.Criminal Behavior and Mental Health, 20(4), 251–256. doi:10.1002/cbm.772
Howells, K., Tennant, A., Day, A., & Elmer, R. (2010). Mindfulness in forensic mental health: Doesit have a role? Mindfulness, 1(1), 4–9. doi:10.1007/s12671-009-0001-x
Ibanez, M. I., Ruiperez, M. A., Moya, J., Marques, M. J., & Ortet, G. (2005). A short version of theSelf-Regulation Inventory (SRI-S). Personality and Individual Differences, 39(6), 1055-1059.doi:10.1016/j.paid.2005.02.029
IBM Corporation (2011). IBM SPSS statistics for windows (Version 20.0) [Software]. Armonk,New York: IBM Corporation.
Inzlicht, M. & Schmeichel, B.J. (2012). What Is ego depletion? Toward a mechanistic revision ofthe resource model of self-control. Perspectives on Psychological Science, 7(5), 450–463.doi:10.1177/1745691612454134
Inzlicht, M., & Gutsell, J. (2007). Running on empty: Neural signals for self-control failure.Psychological Science, 18(11), 933–937. doi:10.1111/j.1467-9280.2007.02004.x
Jacobson, E. (1938). Progressive relaxation. Chicago: University of Chicago Press.
Jain, S., Shapiro, S. L., Swanick, S., Roesch, S. C., Mills, P. J., Bell, I., & Schwartz G. E. R. (2007). Arandomized controlled trial of mindfulness meditation versus relaxation training: Effects ondistress, positive states of mind, rumination, and distraction. Annals of Behavioral Medicine,33(1), 11–21. doi:10.1207/s15324796abm3301_2
Jakupcak, M., Lisak, D., & Roemer, L. (2002). The role of masculine ideology and masculinegender role stress in men’s perception of relationship violence. Psychology of Men andMasculinity, 3(2), 97–106. doi:10.1037/1524-9220.3.2.97
Joireman, J., Anderson, J., & Strathman, A. (2003). The aggression paradox: Understanding linksamong aggression, sensation seeking, and the consideration of future consequences. Journal ofPersonality and Social Psychology, 84(6), 1287–1302. doi:10.1037/0022-3514.84.6.1287
Joireman, J., Balliet, D., Sprott, D., Spangenberg, E., & Shultz, J. (2008). Consideration of futureconsequences, ego-depletion, and self-control: Support for distinguishing between CFC-immediate and CFC-future sub-scales. Personality and Individual Differences, 45(1), 15–21.doi:10.1016/j.paid.2008.02.011
Kabat-Zinn J. (1990). Full catastrophe living: Using the wisdom of your body and mind to facestress, pain, and illness. New York: Dell Publishing.
Kabat-Zinn, J. (1994). Where you go there you are: Mindfulness meditation in everyday life. NewYork: Hyperion.
230
Kabat-Zinn, J. (2003). Mindfulness-based interventions in context: Past, present, and future.Clinical Psychology: Science and Practice, 10(2), 144–156. doi:10.1093/clipsy.bpg016
Kahn, M. M. (2005). Suicide prevention and developing countries. Journal of the Royal Society ofMedicine, 98(10), 459–463. Retrieved from http://jrs.sagepub.com/
Kapur, N., Cooper, J., O'Connor, R. C., & Hawton, K. (2013). Non-suicidal self-injury v. attemptedsuicide: new diagnosis or false dichotomy? British Journal of Psychiatry, 202, 326-328.doi:10.1192/bjp.bp.112.116111.
Kelley, T. M., & Lambert, E. G. (2012). Mindfulness as a potential means of attenuating angerand aggression for prospective criminal justice professionals. Mindfulness, 3(4), 261-274.doi:10.1007/s12671-012-0090-9
Keng, S-L., Smoski, M. J., & Robins, C. J. (2011). Effects of mindfulness on psychological health: Areview of empirical studies. Clinical Psychology Review, 31(6), 1041-1056.doi:10.1016/j.cpr.2011.04.006
Kenny, D. (2013, May 1). Mediation. Retrieved from http://www.davidakenny.net/cm/mediate.htm
Kochanska, G., Coy, K. C. & Murray, K. T. (2001). The development of self-regulation in the firstfour years of life. Child Development, 72(4), 1091–1111. doi:10.1111/1467-8624.00336
Koentjoroningrat. (1985). Javanese Culture. Singapore: Oxford University Press East Asia.
Konrath, S., Bushman, B., & Campbell, W. K. (2006). Attenuating the link between threatenedegotism and aggression. Psychological Science, 17(11), 995–1001. doi:10.1111/j.1467-9280.2006.01818.x
Konrath, S., Bushman, B., & Campbell, W. K. (2006). Attenuating the link between threatenedegotism and aggression. Psychological Science, 17(11), 995–1001. doi:10.1111/j.1467-9280.2006.01818.x
Kramer, R. S. S., Weger, U. W., & Sharma, D. (2013). The effect of mindfulness meditation ontime perception. Consciousness and Cognition, 22(3), 846–852.doi:10.1016/j.concog.2013.05.008
Kraut, R., Olson, J., Banaji, M., Bruckman, A., Cohen, J., & Couper, M. (2004, February/March).Psychological research online: Report of Board of Scientific Affairs' Advisory Group on theConduct of Research on the Internet. American Psychologist, 59(2), 105-117. doi:10.1037/0003-066X.59.2.105
Kross, E., Gard, D., Deldin, P., Clifton, J., & Ayduk, O. (2012). Asking why from a distance: Itscognitive and emotional consequences for people with Major Depressive Disorder. Journal ofAbnormal Psychology, 121(3), 559-569. doi:10.1037/a0028808
Lakey, C. E., Campbell, W. K., Brown, K. W., & Goodie, A. S. (2007). Dispositional mindfulness asa predictor of the severity of gambling outcomes. Personality and Individual Differences, 43(7),1698–1710. doi:10.1016/j.paid.2007.05.007
Lau, M. A., Bishop, S. R., Segal, Z. V., Buis, T., Anderson, N. D., Carlson, L., Shapiro, S., Carmody,J., Abbey, S., & Devins, G. (2006). The Toronto Mindfulness Scale: Development and validation.Journal of Clinical Psychology, 62(12), 1445-1467. doi:10.1002/jclp.20326
231
Lawrence, C. (2006). Measuring individual responses to aggression-triggering events:Development of the Situational Triggers of Aggressive Responses (STAR) scale. AggressiveBehavior, 32(3), 241–252. doi:10.1002/ab.20122
Lawrence, C., & Hodgkins, E. (2009). Personality influences on interpretations of aggressivebehavior: The role of provocation sensitivity and trait aggression. Personality and IndividualDifferences, 46(3), 319-324. doi:10.1016/j.paid.2008.10.022
Lawrence, C., & Hutchinson, L. (2013a). The impact of non-aggressive behaviour early inaggressive interactions: Sex differences in direct and indirect aggression in response toprovocation. British Journal of Psychology. Advance online publication. doi:10.1111/bjop.12020
Lawrence, C., & Hutchinson, L. (2013b). The influence of individual differences in sensitivity toprovocations on provoked aggression. Aggressive Behavior, 39(3), 212–221.doi:10.1002/ab.21473
Li, Y., Wang, M., Wang, C., & Shi, J. (2010). Individualism, collectivism, and Chinese adolescents'aggression: Intracultural variations. Aggressive Behavior, 36(3), 187-94. doi:10.1002/ab.20341
Lieberman, J., Solomon, S., Greenberg, J., & McGregor, H. (1999). A hot new way to measureaggression: Hot sauce allocation. Aggressive Behavior, 25(5), 331–348. doi:10.1002/(SICI)1098-2337(1999)25:5<331::AID-AB2>3.0.CO;2-1
Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline personality disorder. NewYork: Guilford Press.
Linehan, M., M., Bohus, M., & Lynch, T.R. (2007). Dialectical behavior therapy for pervasiveemotion dysregulation. In J.J. Gross (Ed.), Handbook of Emotion Regulation (pp. 581-605). NewYork: Guilford Press.
MacKillop, J., & Anderson, E. J. (2007). Further psychometric validation of the Mindful AttentionAwareness Scale. Journal of Psychopathology and Behavioral Assessment, 29(4), 289–293.doi:10.1007/s10862-007-9045-1
MacKinnon, D. P., & Dwyer, J. H. (1993). Estimating mediated effects in prevention studies.Evaluation Review, 17(2), 144-158. doi:10.1177/0193841X9301700202
MacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2007). Mediation Analysis. Annual Review ofPsychology, 58, 593-614. doi:10.1146/annurev.psych.58.110405.085542
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). Acomparison of methods to test the significance of the mediated effect. Psychological Methods,7(1), 83-104. doi:10.1037/1082-989X.7.1.83
MacKinnon, D. P., Lockwood, M. C., & Williams, J. (2004). Confidence limits for the indirecteffect: distribution of the product and resampling methods. Multivariate Behavioral Research,39(1), 99-128. doi:10.1207/s15327906mbr3901_4
Magnis-Suseno, F. (1997) Javanese ethics and world-view: The Javanese idea of the good life.Jakarta, Indonesia: Penerbit PT Gramedia Pustaka Utama.
Masicampo, E. J., & Baumeister, R. F. (2007). Relating mindfulness and self-regulatory processes.Psychological Inquiry, 18(4), 255-258. doi:10.1080/10478400701598363
232
Maxwell, J.P. (2007). Development and validation of a Chinese version of the Buss–PerryAggression Questionnaire. Journal of Personality Assessment, 88(3), 1–11.doi:10.1080/00223890701317004
Mazlan, N. H., & Ahmad, A. (2012, July - December). The Malay-translated version of theAggression Questionnaire (AQ): The validity and the identification of types of aggression amongfemale prisoners. ASEAN Journal of Psychiatry, 13(2). Retrieved fromhttp://www.aseanjournalofpsychiatry.org/index.php/aseanjournalofpsychiatry/article/viewFile/100/79
McCloskey, M. S., & Berman, M. E. (2003). Alcohol intoxication and self-aggressive behavior.Journal of Abnormal Psychology, 112(2), 306-311. doi:10.1037/0021-843X.112.2.306
Mehta, P. H., & Beer, J. (2010). Neural mechanisms of the testosterone-aggression relation: therole of orbitofrontal cortex. Journal of Cognitive Neuroscience, 22(10), 2357–2368. doi:10.1162/jocn.2009.21389
Metzger, D. S., Koblin, B., Turner, C., Navaline, H., Valenti, E., Holte, S., … Seage, G. R. (2000).Randomized controlled trial of audio computer-assisted self-interviewing: Utility andacceptability in longitudinal studies. American Journal of Epidemiology, 152(2), 99-106.doi:10.1093/aje/152.2.99
Mischkowski, D., Kross, E., & Bushman, B. (2012). Flies on the wall are less aggressive: Self-distancing "in the heat of the moment" reduces aggressive thoughts, angry feelings andaggressive behavior. Journal of Experimental Social Psychology, 48(5), 1187-1191.doi:10.1016/j.jesp.2012.03.012
Moffitt, T., Arseneault, L., Belsky, D., Dickson, N., Hancox, R., Harrington, H. L., … Caspi, A.(2011). A gradient of childhood self-control predicts health, wealth, and public safety.Proceedings of the National Academy of Sciences. doi:10.1073/pnas.1010076108.
Morrison R, & O'Connor, R.C. (2008). A systematic review of the relationship betweenrumination and suicidality. Suicide Life Threat Behavior, 38(5), 523-38.doi:10.1521/suli.2008.38.5.523
Muller, D., Bushman, B.J., Subra, B., & Ceaux, E. (2012). Are people more aggressive when theyare worse off or better off than others? Social Psychological and Personality Science, 3(6), 754-759. doi:10.1177/1948550612436984
Muraven, M. (2010). Building self-control strength: Practicing self-control leads to improvedself-control performance. Journal of Experimental Social Psychology, 46(2), 465–468.doi:10.1016/j.jesp.2009.12.011
Muraven, M., Baumeister, R. F., & Tice, D. M. (1999). Longitudinal improvement of self-regulation through practice: Building self-control through repeated exercise. Journal of SocialPsychology, 139(4), 446–457. doi:10.1080/00224549909598404
Muraven, M., Tice, D. M., & Baumeister, R. F. (1998). Self-control as limited resource: Regulatorydepletion patterns. Journal of Personality and Social Psychology, 74(33), 774–789.doi:10.1037/0022-3514.74.3.774
Nakano, K. (2001). Psychometric evaluation on the Japanese adaptation of the AggressionQuestionnaire. Behaviour Research and Therapy, 39(7), 853–858. doi:10.1016/S0005-7967(00)00057-7
233
National Collaborating Centre for Mental Health. (2012). Self-harm: Longer-term management(Clinical Guideline 133). Retrieved fromhttp://www.nice.org.uk/nicemedia/live/13619/57205/57205.pdf
National Institute for Health and Clinical Excellence (NICE). (2009). Depression in adults: Thetreatment and management of depression in adults (Clinical Guideline 90). Retrieved fromhttp://publications.nice.org.uk/depression-in-adults-cg90
Neave, N., & O'Connor, D. B. (2009, January). Testosterone and male behaviours. ThePsychologist, 22(1), 28-31. Retrieved fromhttp://www.thepsychologist.org.uk/archive/archive_home.cfm?volumeID=22&editionID=171&ArticleID=1455
Nock, M. K. (2010). Self-Injury. Annual Review in Clinical Psychology, 6, 339–63.doi:10.1146/annurev.clinpsy.121208.131258.
Nolen-Hoeksema, S. & Morrow, J. (1991). A prospective study of depression and posttraumaticstress symptoms after a natural disaster: The 1989 Loma Prieta earthquake. Journal ofPersonality and Social Psychology, 61(1), 115-121. doi:10.1037/0022-3514.61.1.115
Novaco, R. W. (2007). Anger dysregulation. In T. Cavell & K. Malcolm (Eds.), Anger, aggression,and interventions for interpersonal violence (pp. 3-54). Mahwah, NJ: Erlbaum.
O’Connor, D. B., Archer, J., & Wu, F. W. C. (2001). Measuring aggression: Self-reports, partnerreports, and responses to provoking scenarios. Aggressive Behavior, 27(2), 79–101.doi:10.1002/ab.2
Oaten, M., & Cheng, K. (2006a). Improved self-control: The benefits of a regular program ofacademic study. Basic and Applied Social Psychology, 28(1), 1–16.doi:10.1207/s15324834basp2801_1
Oaten, M., & Cheng, K. (2006b). Longitudinal gains in self-regulation from regular physicalexercise. British Journal of Health Psychology, 11(4), 717–733. doi:10.1348/135910706X96481
Oaten, M., & Cheng, K. (2007). Improvements in self-control from financial monitoring. Journalof Economic Psychology, 28(4), 487–501. doi:10.1016/j.joep.2006.11.003
Office for National Statistics (2013, February 7). Crime statistics, focus on: Violent crime andsexual offences, 2011/12. Retrieved from http://www.ons.gov.uk/ons/rel/crime-stats/crime-statistics/focus-on-violent-crime/stb-focus-on--violent-crime-and-sexual-offences-2011-12.html
Ortner, C. N. M., Kilner, S. J., & Zelazo, P. D. (2007). Mindfulness meditation and reducedemotional interference on a cognitive task. Motivation and Emotion, 31(4), 271-283.doi:10.1007/s11031-007-9076-7
Oxford Mindfulness Centre (University of Oxford Department of Psychiatry). (2002) Mindfulness-based Cognitive Therapy. Retrieved from http://mbct.co.uk/
Parrott, D. J., & Giancola, P. R. (2007). Addressing “The criterion problem” in the assessment ofaggressive behavior: Development of a new taxonomic system. Aggression and Violent Behavior,12(3), 280-299. doi:10.1016/j.avb.2006.08.002,
234
Parto, M. & Besharat, M. A. (2011). Mindfulness, psychological well-being and psychologicaldistress in adolescents: Assessing the mediating variables and mechanisms of autonomy andself-regulation. Procedia: Social and Behavioral Sciences, 30, 578 – 582.doi:10.1016/j.sbspro.2011.10.112
Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the BarrattImpulsiveness Scale. Journal of Clinical Psychology, 51(6), 768-774. doi:10.1002/1097-4679(199511)51:6<768::AID-JCLP2270510607>3.0.CO;2-1
Pituch, K. A., Whittaker, T. A., & Stapleton, L. M. (2005). A comparison of methods to test formediation in multisite experiments. Multivariate Behavioral Research, 40(1), 1–23.doi:10.1207/s15327906mbr4001_1
Placidi, G. P., Oquendo, M. A., Malone, K. M., Huang, Y. Y., Ellis, S. P., Mann, J. J. (2001).Aggressivity, suicide attempts, and depression: Relationship to cerebrospinal fluid monoaminemetabolite levels. Biological Psychiatry, 50(10), 783–791. doi:10.1016/S0006-3223(01)01170-2
Plutchik, R., & van Praag, H. M. (1989). The measurement of suicidality, aggressivity andimpulsivity. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 13(Suppl.), 23–34.doi:10.1016/0278-5846(89)90107-3
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects insimple mediation models. Behavior Research Methods, Instruments, and Computers, 36(4), 717-731. doi:10.3758/BF03206553
Preacher, K. J., & Hayes, A. F. (2008). Contemporary approaches to assessing mediation incommunication research. In A. F. Hayes, M. D. Slater, & L. B. Snyder (Eds.), The Sage sourcebookof advanced data analysis methods for communication research (pp. 13-54). Thousand Oaks, CA:Sage.
Preacher, K. J., & Hayes, A. F. (2011). INDIRECT for SPSS and SAS [Macro]. Retrieved fromhttp://www.afhayes.com/spss-sas-and-mplus-macros-and-code.html#indirect
Ramirez, J. M. (2007). Justification of aggression in several Asian and European countries withdifferent religious and cultural background. International Journal of Behavioral Development,31(51), 9-15. Retrieved from http://eprints.ucm.es/8417/
Ramirez, J. M., Andreu, J. M., & Fujihara, T. (2001). Cultural and sex differences in aggression: Acomparison between Japanese and Spanish students using two different inventories. AggressiveBehavior, 27(4), 313-322. doi:10.1002/ab.1014
Ritter, D. & Eslea, M. (2005). Hot sauce, toy guns and graffiti: A critical account of currentlaboratory aggression paradigms. Aggressive Behavior, 31(5), 407-419. doi:10.1002/ab.20066
Roaldset, J. O., Bakken, A. M., & Bjørkly, S. (2011). A prospective study of lipids and serotonin asrisk markers of violence and self-harm in acute psychiatric patients. Psychiatry Research, 186,293–299. doi:10.1016/j.psychres.2010.07.029
Roberton, T., Daffern, M., & Bucks, R.S. (2012). Emotion regulation and aggression. Aggressionand Violent Behavior, 17(1), 72–82. doi:10.1016/j.avb.2011.09.006
Samuelson, M., Carmody, J., Kabat-Zinn, J., & Bratt, M. A. (2007). Mindfulness-Based StressReduction in Massachusetts correctional facilities. Prison Journal, 87(2), 254–268.doi:10.1177/0032885507303753
235
Sauer, S., Walach, H., Schmidt, S., Hinterberger, T., Lynch, S., Büssing, A., Kohls, N. (2013).Assessment of Mindfulness: Review on state of the art. Mindfulness, 4(1), 3-17.doi:10.1007/s12671-012-0122-5
Schmeichel, B. J., & Vohs, K. D. (2009). Self-affirmation and self-control: Affirming core valuescounteracts ego depletion. Journal of Personality and Social Psychology, 96(4), 770–782.doi:10.1037/a0014635
Seeley, E. A., & Gardner, W. L. (2003). The “selfless” and self-regulation: The role of chronicother orientation in averting self-regulatory depletion. Self and Identity, 2(2), 103–117.doi:10.1080/15298860309034
Segal, Z. V., Williams, M. G., & Teasdale, J. D. (2012). Mindfulness-based cognitive therapy fordepression (2nd ed.). New York: Guilford Press.
Selby, E. A., Anestis, M. D., & Joiner, T. E., Jr. (2008). Understanding the relationship betweenemotional and behavioral dysregulation: Emotional cascades. Behaviour Research and Therapy,46(5), 593–611. doi:10.1016/j.brat.2008.02.002
Selby, E. A., Franklin, J., Amanda Carson-Wong, A., & Rizvi, S.L. (in press). Emotional cascadesand self-injury: Investigating instability of rumination and negative emotion. Journal of ClinicalPsychology. doi:10.1002/jclp.21966
Shapiro S.L., Oman D., Thoresen, C. E., Plante, T. G., & Flinders, T. (2008). Cultivatingmindfulness: Effects on well-being. Journal of Clinical Psychology, 64(7), 840–862.doi:10.1002/jclp.20491
Shapiro, S. L. & Carlson, L. E. (2009). The art and science of mindfulness: Integrating mindfulnessinto psychology and the helping professions. Washington: American Psychology Association.
Shapiro, S. L., Brown, K. W., Thoresen, C., &, Plante, T. G. (2011). The moderation ofmindfulness-based stress reduction effects by trait mindfulness: Results from a randomizedcontrolled trial. Journal of Clinical Psychology, 67(3), 267—277. doi:10.1002/jclp.20761
Shapiro, S. L., Carlson, L. E., Astin, J. A., & Freedman, B. (2006). Mechanisms of mindfulness.Journal of Clinical Psychology, 62(3), 373–386. doi:10.1002/jclp.20237
Shapiro, S. L., Oman, D., Thoresen, C. E., Plante, T. G., & Flinders, T. (2008). Cultivatingmindfulness: Effects on well-being. Journal of Clinical Psychology, 64(7), 840–862.doi:10.1002/jclp.20491
Shonin, E., Gordon W. V., Slade, K., & Griffiths M. D. (in press). Mindfulness and other Buddhist-derived interventions in correctional settings: A systematic review. Aggression and ViolentBehavior. doi:10.1016/j.avb.2013.01.002.
Shonin, E., Gordon, W.V., & Griffiths, M. D. (2013). Mindfulness-based interventions: Towardsmindful clinical integration. Frontiers in Psychology for Clinical Settings, 4(194).doi:10.3389/fpsyg.2013.00194
Shrout P. E., & Bolger, N. (2002) Mediation in experimental and nonexperimental studies: newprocedures and recommendations. Psychological Methods, 7(4), 422-45. doi:10.1037/1082-989X.7.4.422
Siegel, R. D., Germer, C. K., & Olendzki, A. (2009). Mindfulness: What is it? Where did it comefrom? In F. Didonna (Ed.), Clinical Handbook of Mindfulness (pp. 17-35). New York: Springer.
236
Singh, N. N., Lancioni, G. E., Karazsia, B. T., Winton, A. S. W., Myers, R. E., Singh, A. N. A., …Singh, J. (2013). Mindfulness-based treatment of aggression in individuals with mild intellectualdisabilities: A waiting list control study. Mindfulness, 4(2), 158-167. doi:10.1007/s12671-012-0180-8
Singh, N. N., Lancioni, G. E., Manikam, R., Winton, A. S. W., Singh, A. N. A., Singh, J., & Singh, A.D. A. (2011). A mindfulness-based strategy for self-management of aggressive behavior inadolescents with autism. Research in Autism Spectrum Disorders, 5(3), 1153-1158.doi:10.1016/j.rasd.2010.12.012
Singh, N. N., Lancioni, G. E., Singh Joy, S. D., Winton, A. S. W., Sabaawi, M., Wahler, R. G., &Singh, J. (2007). Adolescents with conduct disorder can be mindful of their aggressive behavior.Journal of Emotional and Behavioral Disorders, 15(1), 56–63.doi:10.1177/10634266070150010601
Singh, N. N., Wahler, R. G., Adkins, A. D., & Myers, R. E. (2003). Soles of the feet: A mindfulness-based self-control intervention for aggression by an individual with mild mental retardation andmental illness. Research in Developmental Disabilities, 24(3), 158–169. doi:10.1016/S0891-4222(03)00026-X
Slee, N. (2008). Cognitive-behavioural therapy for deliberate self-harm (Doctoral dissertation,Universitiet Leiden). Retrieved fromhttps://openaccess.leidenuniv.nl/bitstream/1887/12835/1/166060-bw-Slee.pdf
Slee, N., Garnefski, N., Spinhoven, P., & Arensman, E. (2008). The influence of cognitive emotionregulation strategies and depression severity on deliberate self-harm. Suicide and Life-Threatening Behavior, 38(3), 274-286. doi:10.1521/suli.2008.38.3.274
Slee, N., Spinhoven, P., Garnefski, N., & Arensman, E. (2008). Emotion regulation as mediator oftreatment outcome in therapy for deliberate self-harm. Clinical Psychology Psychotherapy,15(4), 205–216. doi:10.1002/cpp.577
Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equationmodels. Sociological Methodology, 13, 290–312. doi:10.2307/270723
Strathman, A., Gleicher, F., Boninger, D. S., & Edwards, C. S. (1994). The consideration of futureconsequences: Weighing immediate and distant outcomes of behavior. Journal of Personalityand Social Psychology, 66(4), 742–752. doi:10.1037/0022-3514.66.4.742
Stucke, T. S., & Baumeister, R. F. (2006). Ego depletion and aggressive behavior: Is the inhibitionof aggression a limited resource? European Journal of Social Psychology, 36(1), 1–13.doi:10.1002/ejsp.285
Sujiva, V. (2000). Essentials of insight meditation practice: A pragmatic approach to vipassana.Selangor, Malaysia: Buddhist Wisdom Centre.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts goodadjustment, less pathology, better grades, and interpersonal success. Journal of Personality,72(2), 271–322. doi:10.1111/j.0022-3506.2004.00263.x
Taylor, S. (1967). Aggressive behavior and physiological arousal as a function of provocation andthe tendency to inhibit aggression. Journal of Personality, 35(2), 297-310. doi:10.1111/j.1467-6494.1967.tb01430.x
237
Tedeschi J., & Quigley, B. (1996). Limitations of laboratory paradigms for studying aggression.Aggression and Violent Behavior, 1(2), 163–177. doi:10.1016/1359-1789(95)00014-3
Thompson, B. L., & Waltz, J. (2007). Everyday mindfulness and mindfulness meditation:Overlapping constructs or not? Personality and Individual Differences, 43(7), 1875-1885.doi:10.1016/j.paid.2007.06.017
Tice, D. M., Baumeister, R. F., Shmueli, D., & Muraven, M. (2007). Restoring the self: Positiveaffect helps improve self-regulation following ego depletion. Journal of Experimental SocialPsychology, 43(3), 379-384. doi:10.1016/j.jesp.2006.05.007
Tresno, F., Ito, Y., Mearns, J. (2012) Self-Injurious behavior and suicide attempts amongIndonesian college students. Death Studies, 36(7), 627–639.doi:10.1080/07481187.2011.604464
Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview Press.
Twenge, J. M., Baumeister, R. F., Tice, D. M., & Stucke, T. S. (2001). If you can’t join them, beatthem: Effects of social exclusion on aggressive behavior. Journal of Personality and SocialPsychology, 81(6), 1058–1069. doi:10.1037/0022-3514.81.6.1058
Tyler, J. M., & Burns, K. C. (2009). Triggering conservation of the self’s regulatory resources.Basic and Applied Social Psychology, 31(3), 255–266. doi:10.1080/01973530903058490
Tyson, P. D. (1998). Physiological arousal, reactive aggression, and the induction of anincompatible relaxation response. Aggression and Violent Behavior, 3(2), 143–158.doi:10.1016/S1359-1789(97)00002-5
Vago, D. R., & Silbersweig, D.A. (2012). Self-awareness, self-regulation, and self-transcendence(S-ART): A framework for understanding the neurobiological mechanisms ofmindfulness. Frontiers in Human Neuroscience, 6(296). doi:10.3389/fnhum.2012.00296
Vijayakumar, L. (2004). Suicide prevention: The urgent need in developing countries. WorldPsychiatry, 3(3), 158–159. Retrieved fromhttp://www.ncbi.nlm.nih.gov/pmc/articles/PMC1414701/
Vohs, K., Glass, B. D., Maddox, W. T., & Markman, A. B. (2011). Ego depletion is not just fatigue:Evidence from a total sleep deprivation experiment. Social Psychological and PersonalityScience, 2(2), 166-173. doi:10.1177/1948550610386123
Wallace, A. B., & Bodhi, B. (2006). The nature of mindfulness and its role in Buddhist meditation:A correspondence between B. Alan Wallace and the venerable Bhikkhu Bodhi (Unpublishedmanuscript). Santa Barbara, California: Santa Barbara Institute for Consciousness Studies.
Watson, D., Clark, L.A., & Tellegen, A. (1988). Development and validation of brief measures ofpositive and negative affect: the PANAS scales. Journal of Personality and Social Psychology,54(6), 1063-1070, doi:10.1037/0022-3514.54.6.1063.
Westen, D. (2001). Psychology: Mind, brain, and culture, take note! (3rd ed.). New York: Wiley.
Williams, J. M. & Penman, D. (2011). Mindfulness: A practical guide to finding peace in a franticworld. London, England: Piatkus
Williams, J. M. (2010). Mindfulness and psychological process. Emotion, 10(1), 1-17.doi:10.1037/a0018360
238
Williams, J. M., Dugan, D. S., Crane, C., & Fennell, M. J. (2006). Mindfulness-based cognitivetherapy for prevention of recurrence suicidal behaviour. Journal of Clinical Psychology inSession, 62(2), 201–210. doi:10.1002/jclp.20223
World Health Organization. (2008). Suicide and suicide prevention in Asia. Geneva: World HealthOrganization.
Wright, S., Day, A., & Howells, K. (2009). Mindfulness and the treatment of anger problems.Aggression and Violent Behavior, 14(5), 396–401. doi:10.1016/j.avb.2009.06.008.
Wrosch, C., Scheier, M. F., Carver, C. S., & Schulz, R. (2003). The importance of goaldisengagement in adaptive self-regulation: When giving up is beneficial. Self and Identity, 2(1),1–20. doi:10.1080/15298860309021
Wrosch, C., Scheier, M. F., Miller, G. E., Schulz, R., & Carver, C. S. (2003). Adaptive self-regulationof unattainable goals: Goal disengagement, goal reengagement and subjective well-being.Personality and Social Psychology Bulletin, 29(12), 1494–1508. doi:10.1177/0146167203256921
Zhang, P., Roberts, R. E., Liu, Z., Meng, X., Tang, J., Sun, L., & Yu, Y. (2012). Hostility, physicalaggression and trait anger as predictors for suicidal behavior in Chinese adolescents: A school-based study. PLoS One, 7(2), e31044. doi:10.1371/journal.pone.0031044
Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truthsabout mediation analysis. Journal of Consumer Research, 37(August), 197-206.doi:10.1086/651257
Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety And Depression Scale. ActaPsychiatric Scandinavia, 67(3), 361-370. doi:10.1111/j.1600-0447.1983.tb09716.x.
239
APPENDICESAppendix 2.1.Trait measures for Study 1 to 4
(All trait measures were presented online. The same following instruction was used except for the Hospital Anxiety and Depression [HADS]: Zigmond &Snaith, 1983 and the Deliberate Self-Harm Inventory [DSHI]: Gratz, 2001.): This survey asks you a little about yourself. Please indicate how much each of thefollowing statements reflects how you typically are. Everyone feels differently about this so there are no right or wrong answers, we are interested in youropinions. Do not spend too long on any one statement.
Mindful Attention Awareness Scale (MAAS: Brown & Ryan, 2003)
How frequently or infrequently you currently have each experience?AlmostAlways
VeryFrequently
SomewhatFrequently
SomewhatInfrequently
VeryInfrequently
AlmostNever
1. I could be experiencing some emotion and not be conscious of ituntil some time later.
2. I break or spill things because of carelessness, not payingattention, or thinking of something else.
3. I find it difficult to stay focused on what's happening in thepresent.
4. I tend to walk quickly to get where I'm going without payingattention to what I experience along the way.
5. I tend not to notice feelings of physical tension or discomfortuntil they really grab my attention.
6. I forget a person's name almost as soon as I've been told it forthe first time.
7. It seems I am "running on automatic," without much awarenessof what I'm doing.
240
How frequently or infrequently you currently have each experience?AlmostAlways
VeryFrequently
SomewhatFrequently
SomewhatInfrequently
VeryInfrequently
AlmostNever
8. I rush through activities without being really attentive to them.
9. I get so focused on the goal I want to achieve that I lose touchwith what I'm doing right now to get there.
10. I do jobs or tasks automatically, without being aware of whatI'm doing.
11. I find myself listening to someone with one ear, doingsomething else at the same time.
12. I drive places on "automatic pilot" and then wonder why I wentthere.
13. I find myself preoccupied with the future or the past.
14. I find myself doing things without paying attention.
15. I snack without being aware that I'm eating.
Brief Self-Control Scale (Brief SCS: Tangney et al., 2004)
How much each of the following statements reflectshow you typically are?
Not at allModeratelyinaccurate
Neither inaccurateor accurate
Moderatelyaccurate
Verymuch
1. I am good at resisting temptation.
2. I have a hard time breaking bad habits.
3. I am lazy.
4. I say inappropriate things.
5. I do certain things that are bad for me, if they are fun.
6. I refuse things that are bad for me.
7. I wish I had more self-discipline.
8. People would say that I have iron self- discipline.
241
How much each of the following statements reflectshow you typically are?
Not at allModeratelyinaccurate
Neither inaccurateor accurate
Moderatelyaccurate
Verymuch
9. Pleasure and fun sometimes keep me from getting work done.
10. I have trouble concentrating.
11. I am able to work effectively toward long-term goals.
12. Sometimes I can’t stop myself from doing something, even if Iknow it is wrong.
13. I often act without thinking through all the alternatives.
Hospital Anxiety and Depression (HADS: Zigmond & Snaith, 1983)
Read each item and place a firm tick in the box opposite the reply, which comes closest to how you have been feeling IN THE PAST WEEK.1. I feel tense or wound up: 8. I feel as if I am slowed down:
Most of the time Nearly all the time A lot of the time Very often Time to time, occasionally Sometimes Not at all Not at all
2. I still enjoy the things I used to enjoy: 9.I get a sort of frightened feeling like “butterflies” inthe stomach:
Definitely as much Not at all Not quite so much Occasionally Only a little Quite often Hardly at all Very often
3.I get a sort of frightened feeling as if somethingawful is about to happen:
10. I have lost interest in my appearance:
242
Very definitely and quite badly Definitely Yes, but not too badly I don’t take so much care as I should A little, but it doesn’t worry me I may not take quite as much care Not at all I take just as much care as ever
4. 4 I can laugh and see the funny side of things: 11. I feel restless as if I have to be on the move:As much as I always could Very much indeed Not quite so much now Quite a lot Definitely not so much now Not very much Not at all Not at all
5. Worrying thoughts go through my mind: 12. I look forward with enjoyment to things:A great deal of the time As much as I ever did A lot of the time Rather less than I used to From time to time bur not too often Definitely less than I used to Only occasionally Hardly at all
6. I feel cheerful: 13. I get sudden feelings of panic:Not at all Very often indeed Not often Quite often Sometimes Not very often Most of the time Not at all
7. I can sit at ease and feel relaxed: 14. I can enjoy a good book or radio or TV programme:Definitely Often Usually Sometimes Not often Not often Not at all Very seldom
243
Situational Triggers of Aggressive Responses (STAR: Lawrence, 2006)
How accurately each of the following situationsmakes you feel aggressive?
Very inaccurateModeratelyinaccurate
Neither inaccurateor accurate
Moderatelyaccurate
Accurate
1. A friend betrays me.
2. I am the subject of a practical joke.
3. People I live with show a lack of consideration.
4. Someone steals something from me.
5. I feel frustrated.
6. Someone insults me.
7. I have academic or work problems.
8. I experience family dispute.
9. I feel hot and crowded.
10. Someone ignores me.
11. Someone behaves in an inconsiderate manner towards me.
12. I am in pain.
13. I am goaded or provoked by someone.
14. I’ve been let down by someone.
15. I feel stressed.
16. Someone is drunk and behaves inconsiderately towards me.
17. I hear a noise that I cannot control.
18. I am frustrated with services.
19. Others around me are becoming aggressive.
20. Someone makes offensive remarks to me.
21. Another driver commits a traffic violation.
22. I argue with a friend.
244
Aggression Questionnaire (AQ: Buss & Perry, 1992)
How accurately each of the following statements describes the way inwhich you act when you feel angry or aggressive?
Veryinaccurate
Moderatelyinaccurate
Neither inaccurateor accurate
Moderatelyaccurate
Accurate
1. Occasionally I can’t control the urge to strike someone.
2. I tell my friends openly when I disagree with them.
3. I flare up quickly but also get over it quickly.
4. I am sometimes eaten up with jealousy.
5. Given enough provocation I may hit another person.
6. I often find myself disagreeing with people.
7. When frustrated, I let my irritation show.
8. At time I feel I have gotten a raw deal out of life.
9. If somebody hits me, I hit back.
10. When people annoy me, I may tell them what I think.
11. I sometimes feel like I’m about to explode.
12. Other people always seem to get the breaks.
13. I get into fights a little more than the average person.
14. I can’t help getting into arguments when people disagree with me.
15. I am an even-tempered person.
16. I wonder why sometimes I feel so bitter about things.
17. If I have to resort to violence to protect my rights then I will.
18. My friends say that I’m somewhat argumentative.
19. Some of my friends think I’m a hothead.
20. I know that ‘friends’ talk about me behind my back.
21. There are people who pushed me so far we came to blows.
22. Sometimes I fly off the handle for no good reason.
23. I am suspicious of overly friendly strangers.
245
How accurately each of the following statements describes the way inwhich you act when you feel angry or aggressive?
Veryinaccurate
Moderatelyinaccurate
Neither inaccurateor accurate
Moderatelyaccurate
Accurate
24. I can think of no good reason for ever hitting a person .
25. I have trouble controlling my temper.
26. I sometimes feel that people are laughing at me behind my back.
27. I have threatened people that I know.
28. When people are especially nice, I wonder what they want.
29. I have become so mad that I have broken things.
Deliberate Self-Harm Inventory (DSHI: Gratz, 2001)
This questionnaire asks about a number of different things that people sometimes do to hurt themselves. Please be sure to read each question carefullyand respond honestly. Often, people who do these kinds of things to themselves keep it a secret, for a variety of reasons. However, honest responses tothese questions will provide us with greater understanding and knowledge about these behaviours and the best way to help people.
Please answer YES to a question only if you did the behaviour INTENTIONALLY, or on purpose, to hurt yourself; without intending to kill yourself. Do notrespond yes if you did something accidentally (e.g., tripped and banged you head by accident). If you answer yes, then please indicate the FREQUENCY ofeach behaviour.
Have you ever intentionally (i.e., on purpose)… (without intending to kill yourself)Indicate Yes or
No
If Yes, how many times have youdone this?
Less than 5times
More than 5times
1. Cut your wrist, arms, or other area(s) of your body? Y / N
2. Burned yourself with a cigarette? Y / N
3. Burned yourself with a lighter or a match? Y / N
246
Have you ever intentionally (i.e., on purpose)… (without intending to kill yourself)Indicate Yes or
No
If Yes, how many times have youdone this?
Less than 5times
More than 5times
4. Carved words into your skin? Y / N
5. Carved pictures, designs, or other marks into your skin? Y / N
6. Severely scratched yourself, to the extent that scarring or bleeding occurred? Y / N
7. Bit yourself, to the extent that you broke the skin? Y / N
8. Rubbed sandpaper on your body? Y / N
9. Dripped acid onto your skin? Y / N
10. Used bleach, comet, or oven cleaner to scrub your skin? Y / N
11. Stuck sharp objects such as needles, pins, staples, etc. into your skin, not including tattoos,ear piercing, needles used for drug use, or body piercing?
Y / N
12. Rubbed glass into your skin? Y / N
13. Broken your own bones? Y / N
14. Banged your head against something, to the extent that you caused a bruise to appear? Y / N
15. Punched yourself, to the extent that you caused a bruise to appear? Y / N
16. Prevented wounds from healing? Y / N
17. Done anything else to hurt yourself that was not asked about in this questionnaire? If yes,what did you do to hurt yourself? -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Y / N
247
Appendix 2.2.Advertising materials for Study 1
248
Appendix 2.3Information to participants for Study 1
ONLINE SURVEY ON MINDFULNESS, SELF-CONTROL, AGGRESSION, AND SELF-HARM
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
You are invited to take part in a survey on mindfulness, self-control, aggression, and self-harm. Before you decide, it is important for you to understand why the research is beingdone and what it will involve.
This study is not concerned with your actual levels of mindfulness, self-control, aggression,and self-harm – but we are interested in how these behaviours and propensities areassociated with each other. You will be asked to carry out a set of questionnaires whichshould take about 15 minutes.
Please answer AS HONESTLY AS YOU CAN. All information obtained during the study will beCONFIDENTIAL. Any data which we collect will be ANONYMISED.
We really appreciate your help in participating in this study. If at any time you decide thatyou do not want to continue to take part in the study, you are FREE TO WITHDRAW byclicking the "Exit this survey" link at the top right corner of every page.
Optionally, by participating in this study and providing an email address you will be enteredinto a £25 PRIZE DRAW.
If you would like to discuss anything further, please email me – Cleo Yusainy [email protected]. . Alternatively you can call me on 07557199702 or find me in roomB76 in the School of Psychology. You can also contact the supervisor of this study (Dr. ClaireLawrence) at [email protected].
249
Appendix 2.4Electronic consent form for Study 1 and Study 2 (Time 1)
250
Appendix 2.5Debrief materials for Study 1
ONLINE SURVEY ON MINDFULNESS, SELF-CONTROL, AGGRESSION, AND SELF-HARM
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
Thank you for taking part in the study. All data will be kept confidential and used forresearch purposes only. By clicking the 'DONE' button below (and providing an emailaddress) you are also entering a £25 prize draw.
If you have any questions about the issues covered in this study, feel free to contact theResearcher - Cleo Yusainy at [email protected] or on 07557199702. You can alsocontact the supervisor of this study (Dr. Claire Lawrence) [email protected].
USEFUL POINTS OF CONTACT
The following points of contact will also be able to provide you with any information youneed:
Cripps Health Centre (Nottingham students’ and staff health centre): 0115 950 1654
Nightline (University of Nottingham): Offers support, practical advice and information toanyone concerned about mental distress as well as practical information. Tel: 0115 951 4985http://www.su.nottingham.ac.uk/organisation/NottinghamNightline/(also available for non-university participants). Open daily 7pm – 8am.
University of Nottingham Counselling Service: 0115 951 3695
Mind Info Line: Information on all aspects of mental distress. 0845 766 0163—9am to 5pm,Monday to Friday. http://www.mind.org.uk
Samaritans: Free emotional support to anyone going through a crisis. 08457 90 90 90—24hours a day, 7 days a week. http://www.samaritans.org.uk
Harmless: Nottingham based self-harm support group. http://www.harmless.org.uk
251
Appendix 3.1State measures for Study 2, 3, and 4
Positive Affect, Negative Affect Schedule (PANAS: Watson et al., 1988)
This scale consists of a number of words that describe different feelings and emotions. Readeach item and then mark the appropriate answer in the space next to that word. Indicate towhat extent you feel this way right now, that is, at the present moment.
Use the following scale to record your answers.
1 2 3 4 5Very slightly or
not at allA little Moderately Quite a bit Extremely
_____ interested _____ irritable_____ distressed _____ alert_____ excited _____ ashamed_____ upset _____ inspired_____ strong _____ nervous_____ guilty _____ determined_____ scared _____ attentive_____ hostile _____ jittery_____ enthusiastic _____ active_____ proud _____ afraid
252
Toronto Mindfulness Scale (TMS: Lau et al., 2006)
How well does the statement describewhat you just experienced, just now?
Not at all A little Moderately Quite a bit Very much
1. I experienced myself as separate from my changing thoughts andfeelings.
2. I was more concerned with being open to my experiences thancontrolling or changing them.
3. I was curious about what I might learn about myself by takingnotice of how I react to certain thoughts, feelings or sensations.
4. I experienced my thoughts more as events in my mind than as anecessarily accurate reflection of the way things ‘really’ are.
5. I was curious to see what my mind was up to from moment tomoment.
6. I was curious about each of the thoughts and feelings that I washaving.
7. I was receptive to observing unpleasant thoughts and feelingswithout interfering with them.
8. I was more invested in just watching my experiences as they arose,than in figuring out what they could mean.
9. I approached each experience by trying to accept it, no matterwhether it was pleasant or unpleasant.
10. I remained curious about the nature of each experience as it arose.
11. I was aware of my thoughts and feelings without overidentifyingwith them.
12. I was curious about my reactions to things.
13. I was curious about what I might learn about myself by just takingnotice of what my attention gets drawn to.
253
Appendix 3.2Daily remainder for self-control group in Study 2
Hi there,
I hope you are well. This is day-n of your improving posture exercise. To gain the benefit ofan improved posture, it is crucial that you follow this direction and exert as much control aspossible.
To help you track a record on the progress of your exercise, please indicate as honestly aspossible how frequently (on scale 1-6) you comply with the instruction of maintaining goodposture (e.g., sit up straight, walk erectly, etc.) at all time today.
1 = Almost Never2 = Very Infrequently3 = Somewhat Infrequently4 = Somewhat Frequently5 = Very Frequently6 = Almost Always
You can do this by:a. Simply replying this email by writing down your compliance on scale 1 to 6, or
alternatively:b. Downloading the ‘Improving Posture Diary’ document (attached) and filling out your
compliance score on today’s column only. If you prefer this option, you would onlyneed to download this document once. Please keep it in a convenient place for youto fill it in every evening for two-week period.
To help you keep the track on your progress, we will send you a daily reminder about theexercise again tomorrow.
If you would like to discuss anything further, please email me – Cleo [email protected]. Alternatively you can call me on 07557199702 or find me in roomB76 in the School of Psychology. You can also contact the supervisor of this study (Dr. ClaireLawrence) at [email protected].
We really appreciate your help in participating in our study. Participation is totally voluntary.All data collected will be kept confidential.
Kind regards,Cleo Yusainy
254
Appendix 3.3Mindfulness instruction for Study 2
Mindfulness walking exercise (Kabat-Zinn, 1994; Sujiva, 2000)
The first task that you need to do today is called a mindfulness exercise. Mindfulness meansfully experiencing what is happening in and around at this very moment. To facilitate thisstate of mindfulness, you will be asked to do a mindfulness walking exercise. This involvesintentionally attending to the experience of walking itself; in other words it means simplywalking and knowing that you are walking. Mindfulness walking is similar to normal walking,only more deliberate, relax, and slower—but not too slowly that you cannot feel anysensations. To help you make sense of this exercise, I will present you a video demonstrationfirst. (VIDEO)
Before you do this exercise, there are some points that you need to bear in mind:
Since the primary purpose of the mindfulness walking is to help you become aware ofthe present moment, you need to say every mental note silently while you are doing it,not before or afterwards.
Your attention should be directed towards witnessing the full experience of walking (themovements, shifts of weight and balance, and sensations in the feet and legs associatedwith walking), as opposed to the physical feet per se. If you begin to notice your mindwandering, mentally take a note of whatever emotions, thoughts, feelings etc. arise.There is no need to avoid these emotions or to become occupied with them, simplyobserve what is happening. If however your mind gets too distracted, you may stopwalking. Notice how these distractions would eventually pass away; gently bring yourattention back to the intention of walking, then continue to walk again.
There is no right or wrong in this exercise. You may do this with your shoes on, or youmake take them off if this would help you feel more comfortable.
Now you will be guided to do this exercise as the narrator says the mental notes aloud forthe first time. Then you will be asked to continue the processes on your own until you areinstructed to stop.1. Begin by standing, your hands in front of your body, one hand clasping the other. Look
straight ahead or toward the floor several feet in front of you.2. Observe the standing posture first. As you do, say in your mind "standing, standing".
Remember that you need to say each mental note while you are doing it, not before orafterwards.
3. Next, note the intention to move. Then lift the right foot and place it down, then the leftfoot. Say in your mind "placing, placing" as you move, not before or afterwards.
4. On the last step, place your feet together, noting "stopping."5. Observe the standing posture again. "Standing ".6. Now you will begin to turn. The turn has four steps. Say in your mind "intending to turn."7. Lift the toes of your right foot and pivot on your heel. Lift the left foot and place it down,
the left foot doesn't pivot, but steps. Say in your mind "turning, turning" as you turn, notbefore or afterwards.
8. Finally note “Intending to walk”, and continue walking.9. Follow the movement with your mind, don’t look at your feet.10. If your mind gets too distracted, you may stop walking. Notice how these distractions
would pass away. Then bring your attention back to the walking.11. Now you may stop walking. Stand still and give yourself a few moments to comprehend
what you have just experienced before you call the researcher.
255
Appendix 3.4Relaxation instruction for Study 2
Progressive Muscle Relaxation (PMR: Jacobson, 1938)
The first task that we will have you do today is called a progressive muscle relaxation.Progressive muscle relaxation is an exercise that will help you relax your mind and body byprogressively tensing and relaxing muscle groups throughout your entire body. You willtense each muscle group vigorously, but without straining, and then suddenly release thetension and feel the muscle relax. To help you make sense of this exercise, I will present youa video demonstration first. (VIDEO)
Before you do this exercise, there are some points that you need to bear in mind:
You will tense each muscle for about 5 seconds. If you have any pain or discomfort inany of the targeted muscle groups feel free to omit that step. Throughout this exerciseyou may visualize the muscles tensing and a wave of relaxation flowing over them as yourelease that tension.
Allow your attention to focus only on your body. If you begin to notice your mindwandering, bring it back to the muscle you are working on. It is important that you keepbreathing throughout the exercise.
There is no right or wrong in this exercise. You may do this with your shoes on, or youmake take them off if this would help you feel more comfortable.
Now you will be guided to do this exercise as the narrator says the mental notes. Then youwill be asked to continue the processes on your own until you are instructed to stop.1. Let’s start by finding a comfortable sitting position in your chair, with both feet planted
on the floor and your hands resting loosely in your lap. You can close your eyes if todeepen the relaxation you feel comfortable doing so.
2. Now take a few slow deep breaths. Draw the air all away to the bottom of your lungs sothat you feel your abdomen expand with each inhalation.
3. Allow your body to relax with each breath and try to give yourself permission to leave allyour worries behind for the next several minutes. Simply focus on your body.
4. As you continue your slow deep breathing, shift your focus to your feet. Curl your toesunder tightly, tightly, and hold that tension… and then release that tension all at once,and let your feet go limp. Pay attention to the tingling sensation as the tension flows outof your feet.
5. Now press your knees together hard and hold that tension, hold it, hold it… and nowrelease, and feel the relaxation spreads through your legs.
6. Take a moment to enjoy the warm sensation, and return your attention to your slowdeep breathing.
7. Now tighten your thighs and buttocks by pressing your heels downwards firmly to thefloor, hold this, hold this … and then relax. Feel your relaxation deepened, as the tensiondrains from your body, and your body grows heavier in your chair.
8. Now breath in deeply and a fill up your lungs completely. Hold your breath, experiencethat tension in your chest … now exhale, and let your chest relax. Continue breathing,letting your breath come freely and gently. Imagine the tension draining from your bodywith each exhalation.
9. Now shrug your shoulders up as if they could touch your ears, hold that tension, tighter,tighter … and release. Feel the relaxation spread through your neck and shoulders,melting away any tension that was there.
256
10. Now bend your elbows and tense your biceps as hard as you can, tense it … and release.Feel the tension being replaced by warmth and relaxation.
11. Clench your fists and notice the tension in your fists and forearms … Now relax and dropyour hands back into your lap.
12. Now open your hands wide and stretch your fingers outwards, stretch, stretch … andrelease. Feel the looseness and warmth in your hands.
13. Return your attention to your breath. Breathe slowly and deeply, releasing tension asyou exhale.
14. Now shift your attention to your face. Scrunch up your entire face. Pull your eyebrowsinto a frown, squeeze your eyes shut, clinch your jaw, pucker your lips, hold that, holdthat … now release. Notice the relaxation melt the tension in your face.
15. Continue to breathe slowly and deeply.16. Scan your body from your toes all the way up to your head. Notice areas that still hold
some tension. Return your attention to those area, and tense, and relax once again.17. Now take a few moments to just feel the warmth, heaviness, and comfort of your
relaxed body.18. When you are ready to end your relaxation, gently begin to wiggle your toes and fingers,
give your body a slow gentle stretch before you stand up and call the researcher.
257
Appendix 3.5Advertising materials for Study 2
Participants needed in a Reaction-Time Competition (Two-part study)
Hi there,
You are invited to take part in a study investigating the effects of personality traits andexperimental treatments on the way people perform in a competitive reaction-time task. Ifyou consent to participate in this study, you will be asked to take part in 2 (TWO) TIMESPOINT.
At the first part of the study, you will be asked to complete an ONLINE SURVEY (which willtake only 10 minutes). By completing this part of the study, you will be entered into a £25prize draw.
Based on random allocation, you may or may not be asked to perform a simple daily exerciseof maintaining a good posture on your own for two weeks.
At the second part of the study, you will be asked to play an hour COMPUTER-BASEDCOMPETITIVE REACTION-TIME against an opponent in a laboratory in the School ofPsychology (University Park). By participating in this part, you will be given an inconvenienceallowance of £5*. In addition, a prize of £25 will also be given for fastest reaction time acrossall participants.
Participation in this study is totally voluntary. All data collected will be kept confidential andused for research purposes only.
If you are interested to participate, please email me, - Cleo Yusainy [email protected], with the information about your sex (Male / Female) foradministration procedure.
Thank you,Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
* Alternatively, experimental credits will be given for participants signing up via the ResearchParticipation Scheme (RPS).
258
Appendix 3.6Information to participants for Study 2 (Time 1)
INFORMATION TO PARTICIPANTS REACTION-TIME COMPETITION
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
You are invited to take part in a study looking at PERFORMANCE IN A COMPETITIVEREACTION-TIME GAME. If you consent to participate in this study, you will be asked to takepart in 2 TIMES POINT.
THIS IS PART 1 OF THE STUDY. Questions will be asked about your traits mindfulness, self-control, aggression, and self-harm. Completing this survey should take you about 10MINUTES. By completing this part, you will be entered into a £25 prize draw.
(Instruction in this paragraph was given to self-control training group only.) Afterwards youwill be asked to perform a simple daily exercise of maintaining good posture on your own fora two-week period; and to record your progress in a one-question diary questionnaire eachevening. The purpose of this exercise is to help you improve your posture.
Next, you will be emailed a reminder about Part 2, an hour competitive reaction-time gamein a laboratory in the School of Psychology (University Park). By participating in Part 2, youwill be given an inconvenience allowance of £5*. In addition, a prize of £25 will also be givenfor fastest reaction time across all participants.
Your lack of continuation in Part 2 will not affect your entry into the incentive prize draw.Please note that all payment will be given after Part 2 is completed.
If at any time you decide that you do not want to continue, you are free to withdraw. Alldata collected will be kept confidential and used for research purposes only.
If you would like to discuss anything further, please email me – Cleo Yusainy [email protected]. Alternatively you can contact me on 07557199702 or find me inroom B76 in the School of Psychology. You can also contact the supervisor of this study (Dr.Claire Lawrence) at [email protected].
*Experimental credits will be given for participants signing up via the Research ParticipationScheme (RPS).
259
Appendix 3.7Debrief materials for Study 2 (Time 1)
DEBRIEF MATERIALS REACTION-TIME COMPETITION
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
Thank you very much for taking part in Part 1 of the study. This survey is not concern withthe actual levels of your traits of mindfulness, self-control, aggression and self-harm, butwith how these behaviours and propensities are associated with each other. The emailaddress you provided will be entered into a £25 prize draw.
(Instruction in this paragraph was given to self-control training group only.) Next, we willemail you about your following task, a simple daily exercise to improve your posture (e.g., situp straight, walk erectly, etc.) on your own for two weeks period. Daily reminders will besent via email so that your progress can be reported in a ONE-QUESTION QUESTIONNAIREeach evening.
Then you will be emailed about Part 2, a competitive reaction-time game, so that you canset up a convenient time to come to a laboratory in the School of Psychology (UniversityPark). By participating in Part 2, you will be given an inconvenience allowance of £5*. Inaddition, a prize of £25 will also be given for fastest reaction time across all participants.
If at any time you decide that you do not want to continue, you are free to withdraw. Alldata collected will be kept confidential and used for research purposes only. Please notethat all payment will be given after Part 2 is completed. However your lack of participation inPart 2 will not affect your entry into the incentive prize draw from Part 1.
If you would like to discuss anything further, please email me – Cleo Yusainy [email protected]. Alternatively you can contact me on 07557199702 or find me inroom B76 in the School of Psychology. You can also contact the supervisor of this study (Dr.Claire Lawrence) at [email protected].
*Experimental credits will be given for participants signing up via the Research ParticipationScheme (RPS).
260
Appendix 3.8Information to participants for Study 2 (Time 2), and Study 3 and 4
INFORMATION TO PARTICIPANTS REACTION-TIME COMPETITION
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
You are invited to take part in a study investigating people performance in a reaction-timecompetition.
In the competition, you will be playing reaction-time task against a participant located in adifferent room. Your job is to RESPOND AS FAST AS YOU CAN to the stimuli presented on thecomputer screen. Each time you win a trial, you have the option to deliver a brief (50 ms)noise blast to your opponent. If you lose, your opponent will be given this opportunity. Thenoise blasts you will hear will always be at a clinically safe limit, but if you have any hearingproblems you should let me know immediately and you will not be asked to continue.
Apart from the competition, you will be asked to complete a short survey on traitsmindfulness, self-control, aggression, and self–harm; and to perform some simple unrelatedtasks.
We really appreciate your help in participating in this study. If at any time you decide thatyou do not want to continue to take part, you are free to withdraw. All data collected will bekept confidential and used for research purposes only. Your name or any identifyingcharacteristics will not be available to anyone other than the researcher, at any point.
By completing this study, you will be given an inconvenience allowance of £5*. In addition, aprize of £50 will also be given for fastest reaction time across all participants.
*Experimental credits will be given for participants signing up via the Research ParticipationScheme (RPS).
261
Appendix 3.9Consent form for Study 2 (Time 2), and Study 3 and 4
CONSENT FORM REACTION-TIME COMPETITION
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
The participant should complete the whole of this sheet himself/herself. Please cross out asnecessary:
Have you read and understood the participant information sheet YES/NO
Have you had the opportunity to ask questions and discuss the study YES/NO
Have all the questions been answered satisfactorily YES/NO
Have you received enough information about the study YES/NO
Do you understand that you are free to withdraw from the study:at any time YES/NOwithout having to give a reason YES/NO
Do you agree to take part in the study YES/NO
“This study has been explained to me to my satisfaction, and I agree to take part.I understand that I am free to withdraw at any time.”
Signature of the Participant : Date:Name (in block capitals) :Age :
I have explained the study to the above participant and he/she has agreed to take part.Signature of researcher : Date:Cleo Yusainy
262
Appendix 3.10Debrief materials for Study 2 (Time 2), and Study 3 and 4
DEBRIEF MATERIALS REACTION-TIME COMPETITION
Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
Thank you for taking part of this study. The aim of this study is to investigate the effects ofcertain personality traits and experimental treatments on the way people perform in areaction-time competition.
In the competition, you were in a condition which meant that you were actually competingagainst the computer. The level of noise that the computer gave to you was manipulatedprior to the study. Please refrain from telling any possible future participants about thenature of this study, as they may be in a different condition to the one you did.
By completing this part of the study, you will be given an inconvenience allowance of £5*. Inaddition, a prize of £50 will also be given for fastest reaction time across all.
Your email and data will never be kept in the same place – and so cannot be matched up byanyone other than the researcher. Your responses are totally confidential and are used forresearch purposes only.
If you would like to discuss anything further, please email me – Cleo Yusainy [email protected]. Alternatively you can contact me on 07557199702 or find me inroom B76 in the School of Psychology. You can also contact the supervisor of this study (Dr.Claire Lawrence) at [email protected].
263
Appendix 4.1Mindfulness instruction for Study 3 and 4
Mindfulness of breathing exercise (Williams & Penman, 2011)
For the next several minutes, you will be guided with an audio instruction to perform asimple exercise to induce a particular kind of awareness, called mindfulness of breathing.Mindfulness is paying attention in the present moment, with openness and curiosity, insteadof judgment. We often focus on things other than what is happening in the moment—worrying about the future, thinking about the past, focusing on what is coming next, ratherthan what is right in front of us.
Sometimes we do pay close attention to what we are thinking and feeling, and we becomevery critical of our thoughts and feelings, and we try to either change them, or distractourselves because this critical awareness can be very painful.
Being mindful falls between these two extremes—we pay attention to what is happeninginside and around us, we see events and experiences as what they are, and we allow thingswe can’t control to be as they are, while we focus our attention on the task at hand.
Mindfulness is a process: We do not reach a final and total state of mindfulness. It is a wayof being in one moment that comes and goes.
The best way to understand mindfulness is to practice it, so let’s do that now.1. Settle into a comfortable sitting position, allow your back to adopt an erect, dignified
posture; neither stiff nor tensed up, but comfortable. If sitting on a chair, have your feetflat on the floor with your legs uncrossed. Allow your eyes to close if that feelscomfortable. If not, lower your gaze so it falls, unfocused, a few feet in front of you.
2. Bring your awareness to physical sensations by focusing your attention on thesensations of touch in the body where it is in contact with the floor and with whateveryou are sitting on. Spend a few moments exploring these sensations.
3. Now focusing your attention on your feet, starting with the toes, expand the “spotlightof attention” so it takes in the soles of your feet, the heels and the top of your feet, untilyou are attending to any and all of the physical sensations you become aware of in bothfeet, moment by moment. Spend a few moments attending to the feet in this way,noticing how sensations arise and dissolve in awareness. If there are no sensations inthis region of the body, simply register a blank. This is perfectly fine–you are not tryingto make sensations happen–you are simply registering what is already here.
4. Now, expand your attention to take in the rest of both legs for a few moments, then thetorso (from the pelvis and hips up to the shoulders); then the left arm; then the rightarm; then the neck and head.
5. Spend a minute or two resting in the awareness of the whole body. See if it is possible toallow your body and its sensations to be just as you find them.
6. Now bring your awareness to the breath as it moves in and out of the body at theabdomen. Notice the changing patterns of physical sensations in this region of the bodyas the breath moves in and out. It may help to place your hand here for a few breaths,and feel the abdomen rising and falling.
7. You may notice mild sensations of stretching as the abdomen gently rises with each in-breath, and different sensations as the abdomen falls with each out-breath.
8. As best you can, follow closely with your attention, so you notice the changing physicalsensations for the full duration of each in-breath and the full duration of each out-
264
breath, perhaps noticing the slight pauses between one in-breath and the following out-breath, and between one out-breath and the following in-breath.
9. There is no need to try to control your breathing in anyway at all–simply let the breathbreathe itself.
10. Sooner or later, your attention will wander away from the breath. Simply acknowledgewhere the mind had wandered to. Then gently escort your attention back to the breath.
11. The mind will likely wander over and over again, so each time, remember that the aim issimply to note where the mind has been, then gently escort your attention back to thebreath.
12. Continue with the practice, perhaps reminding yourself from time to time that theintention is simply to be aware of your experience in each moment. Use your breath asan anchor to gently reconnect with the here and now each time that you notice thatyour mind has wandered and is no longer in touch with where you had intended it to be.
13. The sound of the bell will indicate when you should finish your practice. Give yourself afew moments to comprehend what you have just experienced before you call theresearcher.
265
Appendix 4.2Advertising materials for Study 3 and 4 (trans.)
Participate in a Reaction-Time Competition and earn £5 plus the chance to win £50 -NATIVE UK STUDENTS ONLY!
Hi there,
If you are a NATIVE UK student (speak British English as your first language) then you areinvited to take part in a study looking at the effects of personality traits and experimentaltreatments on performance in a fun reaction-time competition.*
In this study, you will be asked to play a reaction-time competition in a laboratory in theSchool of Psychology, University of Nottingham (University Park), and to perform a fewsimple unrelated tasks. For 1 hour of your time, you will be given an inconvenienceallowance of £5, plus chance to win £50 for fastest reaction time across all participants.
Participation is voluntary, confidential, and greatly appreciated. If you are interested, pleaseemail me – Cleo Yusainy at [email protected], with the information about your sex(Male / Female) for administration procedure.
Thank you,Researcher : Cleo YusainySupervisor : Dr. Claire LawrenceSchool of Psychology, University of Nottingham
* Please refrain from participating if you are not a native UK student; or if you haveparticipated in the similar reaction-time competition (game) involving a noise blast in theSchool of Psychology.
266
Appendix 6.1Summary of findings for cross-cultural analysis
Dependent measure British (Chapter 4) Indonesian (Chapter 5) Cross-cultural (Chapter 6)
Blast intensity on noprovocation, low and highprovocations
Mindfulness reducedaggression for depletedparticipants under no andlow provocations.
Depleted individuals weremore aggressive under noand low provocations.
Mindfulness reduced aggressionfor depleted participants underlow and high provocations.
Depleted individuals were moreaggressive under low provocation.
Mindfulness reduced aggression fordepleted participants under all provocationlevels.
Depleted individuals and those not receivingmindfulness induction were more aggressiveunder no and low provocations.
Indonesian male and female participantswere more aggressive under all provocationlevels.
Mindfulness reduced aggression fordepleted females except on highprovocation.
Maximum blast latency Depleted individuals whoreceived mindfulnessinduction delivered earliermaximum blasts.
Females delivered earliermaximum blasts.
Depleted females whoreceived mindfulnessinduction delivered earliermaximum blasts.
Depleted individuals who receivedmindfulness induction deliveredlater maximum blasts.
Females delivered later maximumblasts.
Depleted females who receivedmindfulness induction deliveredlater maximum blasts.
Depleted British participants who receivedmindfulness induction delivered latermaximum blasts.
Females delivered earlier maximum blasts.
267
Dependent measure British (Chapter 4) Indonesian (Chapter 5) Cross-cultural (Chapter 6)
Indirect aggression Mindfulness did notmoderate the link betweendepletion and ratings ofindirect aggression.
Mindfulness did not moderate thelink between depletion and ratingsof indirect aggression.
Indonesians rated the opponent as beingmore aggressive, more fair and lesscompetitive.
Indonesian males and females rated theopponent as being more aggressive but lesscompetitive, only Indonesian females ratedthe opponent as being more fair.
Self-control performance(handgrip stamina) pre-provocation procedure
Positive main effect ofmindfulness induction onself-control performance.
Females outperformedmales.
Mindfulness moderated the linkbetween depletion and self-controlperformance.
Positive main effect of mindfulnessinduction and negative effect ofdepletion condition on self-controlperformance.
Positive main effect of mindfulness inductionand negative effect of depletion condition onself-control performance across cultures.
Indonesian participants performed better.
Indonesian males outperformed their Britishcounterparts.
Self-control performance(handgrip stamina) post-provocation procedure
Positive main effect ofmindfulness induction onself-control performance.
Positive main effect of mindfulnessinduction and negative effect ofdepletion condition on self-controlperformance.
Females outperformed males.
Positive main effect of mindfulness inductionand negative effect of depletion condition onself-control performance across cultures.